Transcription functions as a pivotal biological process in cell biology, which is required to complete the binding of transcription factors (TFs) to transcription factor binding sites (TFBSs) on the DNA. Accurate prediction of TFBSs can provide great potential to regulate the expression of interested genes, which can facilitate exploration of new drugs and treatment for diseases. Although many deep learning-based models have been proposed for predicting TFBSs, existing models still have problems, including the use of convolutional processing of DNA sequences that loses information about the DNA double helix structure and fails to adequately account for the stereoscopic structure of DNA shape data in three dimensions. In this paper, we propose a novel model called DeepCTMS, in which both sequence features and shape features of DNA slices are effectively fused to derive high-quality representations for the task of TFBS prediction. A sequence feature processing module is first used to extract the DNA double helix structure features of DNA slices. The three-dimensional features of DNA shape data are extracted by employing a convolutional triple attention (CTA) module for the shape data of a DNA slice. Finally, a multi-scale fusion feature processing (MSFFP) module is used to fuse sequence features and shape features to obtain representations with significantly aligned semantics of both features. Ablation experiments, t-SNE visual analysis, and cross-cell line validation results demonstrate that DeepCTMS consistently outperforms benchmark models on prediction performance and generalization ability on 165 ChIP-seq datasets.

MOTIVATION: The widespread use of antibiotics has led to the emergence of resistant pathogens. Antimicrobial peptides (AMPs) combat bacterial infections by disrupting the integrity of cell membranes, making it challenging for bacteria to develop resistance. Consequently, AMPs offer a promising solution to addressing antibiotic resistance. However, the limited availability of natural AMPs cannot meet the growing demand. While deep learning technologies have advanced AMP generation, conventional models often lack stability and may introduce unforeseen side effects. RESULTS: This study presents a novel denoising VAE-based model guided by desirable physicochemical properties for AMP generation. The model integrates key features (e.g. molecular weight, isoelectric point, hydrophobicity, etc.), and employs position encoding along with a Transformer architecture to enhance generation accuracy. A customized loss function, combining reconstruction loss, KL divergence, and property preserving loss ensure effective model training. Additionally, the model incorporates a denoising mechanism, enabling it to learn from perturbed inputs, thus maintaining performance under limited training data. Experimental results demonstrate that the proposed model can generate AMPs with desirable functional properties, offering a viable approach for AMP design and analysis, which ultimately contributes to the fight against antibiotic resistance. AVAILABILITY AND IMPLEMENTATION: The data and source codes are available both in GitHub (https://github.com/David-WZhao/PPGC-DVAE) and Zenodo (DOI 10.5281/zenodo.14730711).

Antimicrobial peptides (AMPs) are crucial in addressing the global crisis of bacterial resistance. However, there are still significant limitations in existing methods on de novo AMPs design, especially in designing AMPs with desirable physicochemical properties for specific bacterial pathogens. In this study, we propose a novel generative framework for designing pathogen-targeted antimicrobial peptides with programmable physicochemical properties. More specifically, a conditional Variational Autoencoder is first pretrained for generating AMPs with editable physicochemical properties. We then develop a conditional diffusion model to learn hidden representations of AMPs for targeting pathogens of interest, and construct corresponding MIC predictors for specific bacterial strains. Through comprehensive simulation experiments, we demonstrate that the proposed framework outperforms most existing models in terms of antimicrobial efficacy against specific bacterial targets. Moreover, through systematic screening and analysis, we have identified two star AMPs for each of the two target bacterial species (i.e., E. coli or S. aureus), both of which exhibit excellent performance in antibacterial activity, hemolytic properties, toxicity profiles, etc. Overall, this study provides the key technological support for developing next-generation intelligent platforms for antimicrobial agents design.

Multi-modal multi-view graph learning models have achieved significant success in medical outcome prediction, combining various modalities to enhance the performance of various medical tasks. However, current architectures for multi-modal multi-view graph learning (M3GL) models heavily depend on manual design, demanding significant effort and expert experience. Meanwhile, significant advancements have been achieved in the field of graph neural architecture search (GNAS), contributing to the automated design of learning architectures based on graphs. However, GNAS faces challenges in automating multimodal multi-view graph learning (M3GL) models, as existing frameworks cannot handle M3GL architecture topology, and current search spaces do not consider M3GL models. To address the above challenges, we propose, for the first time, a multi-modal multi-view graph neural architecture search (M3GNAS) framework that automates the construction of the optimal M3GL models, enabling the integration of multi-modal features from different views. We also design an effective multi-modal multi-view learning (M3L) search space to develop inner-view and outer-view graph representation learning in the context of graph learning, obtaining a latent graph representation tailored to the specific requirements of downstream tasks. To examine the effectiveness of M3GNAS, it is evaluated on medical outcome prediction tasks. The experimental findings demonstrate our proposed framework’s superior performance compared to state-of-the-art models.

MOTIVATION: The crisis of antibiotic resistance, which causes antibiotics used to treat bacterial infections to become less effective, has emerged as one of the foremost challenges to public health. Identifying the properties of antibiotic resistance genes (ARGs) is an essential way to mitigate this issue. Although numerous methods have been proposed for this task, most of these approaches concentrate solely on predicting antibiotic class, disregarding other important properties of ARGs. In addition, existing methods for simultaneously predicting multiple properties of ARGs fail to account for the causal relationships among these properties, limiting the predictive performance. RESULTS: In this study, we propose a causality-guided framework for annotating properties of ARGs, in which causal inference is utilized for representation learning. More specifically, the hidden biological patterns determining the properties of ARGs are described by a Gaussian Mixture Model, and procedure of causal representation learning is used to derive the hidden features. In addition, a causal graph among different properties is constructed to capture the causal relationships among properties of ARGs, which is integrated into the task of annotating properties of ARGs. The experimental results on a real-world dataset demonstrate the effectiveness of the proposed framework on the task of annotating properties of ARGs. AVAILABILITY AND IMPLEMENTATION: The data and source codes are available in GitHub at https://github.com/David-WZhao/CausalARG.

The crisis of antibiotic resistance has become a significant global threat to human health. Understanding properties of antibiotic resistance genes (ARGs) is the first step to mitigate this issue. Although many methods have been proposed for predicting properties of ARGs, most of these methods focus only on predicting antibiotic classes, while ignoring other properties of ARGs, such as resistance mechanisms and transferability. However, acquiring all of these properties of ARGs can help researchers gain a more comprehensive understanding of the essence of antibiotic resistance, which will facilitate the development of antibiotics. In this paper, the task of predicting properties of ARGs is modeled as a multi-task learning problem, and an effective subtask-aware representation learning-based framework is proposed accordingly. More specifically, property-specific expert networks and shared expert networks are utilized respectively to learn subtask-specific features for each subtask and shared features among different subtasks. In addition, a gating-controlled mechanism is employed to dynamically allocate weights to subtask-specific semantics and shared semantics obtained respectively from property-specific expert networks and shared expert networks, thus adjusting distinctive contributions of subtask-specific features and shared features to achieve optimal performance for each subtask simultaneously. Extensive experiments are conducted on publicly available data, and experimental results demonstrate the effectiveness of the proposed framework on the task of ARGs properties prediction.

Over the last decade, there has been a significant increase in supporting reproducible computational research (RCR) [1]. The global adoption of the FAIR principles [2] stands as a key indicator of this trend. Specifically, federal and global research funding agencies have increasingly mandated scientific data and related products, such as code and algorithms, be made Findable, Accessible, Interoperable, and Reusable (FAIR) [2].

Recent developments in Large Language Models (LLMs) have advanced the natural language processing (NLP) studies to a new era [1], [2], [4]–[6]. In generic domains, LLMs have become a key component in wide variety of state-of-the-art NLP tasks. In addition, prompt learning enables LLMs-based models to reach robust performance with much smaller training data.

The First Workshop on Personalized Generative AI1 aims to be a cornerstone event fostering innovation and collaboration in the dynamic field of personalized AI. Leveraging the potent capabilities of Large Language Models (LLMs) to enhance user experiences with tailored responses and recommendations, the workshop is designed to address a range of pressing challenges including knowledge gap bridging, hallucination mitigation, and efficiency optimization in handling extensive user profiles. As a nexus for academics and industry professionals, the event promises rich discussions on a plethora of topics such as the development and fine-tuning of foundational models, strategies for multi-modal personalization, and the imperative ethical and privacy considerations in LLM deployment. Through a curated series of keynote speeches, insightful panel discussions, and hands-on sessions, the workshop aspires to be a catalyst in the development of more precise, contextually relevant, and user-centric AI systems. It aims to foster a landscape where generative AI systems are not only responsive but also anticipatory of individual user needs, marking a significant stride in personalized experiences.

Drug-drug interactions (DDIs) are compound effects when patients take two or more drugs at the same time, which may weaken the efficacy of drugs or cause unexpected side effects. Thus, accurately predicting DDIs is of great significance for the drug development and the drug safety surveillance. Although many methods have been proposed for the task, the biological knowledge related to DDIs is not fully utilized and the complex semantics among drug-related biological entities are not effectively captured in existing methods, leading to suboptimal performance. Moreover, the lack of interpretability for the predicted results also limits the wide application of existing methods for DDIs prediction. In this study, we propose a novel framework for predicting DDIs with interpretability. Specifically, we construct a heterogeneous information network (HIN) by explicitly utilizing the biological knowledge related to the procedure of inducing DDIs. To capture the complex semantics in HIN, a meta-path-based information fusion mechanism is proposed to learn high-quality representations of drugs. In addition, an attention mechanism is designed to combine semantic information obtained from meta-paths with different lengths to obtain final representations of drugs for DDIs prediction. Comprehensive experiments are conducted on 2410 approved drugs, and the results of predictive performance comparison show that our proposed framework outperforms selected representative baselines on the task of DDIs prediction. The results of ablation study and cold-start scenario indicate that the meta-path-based information fusion mechanism red is beneficial for capturing the complex semantics among drug-related biological entities. Moreover, the results of case study demonstrate that the designed attention mechanism is able to provide partial interpretability for the predicted DDIs. Therefore, the proposed method will be a feasible solution to the task of predicting DDIs.

It is significant to comprehend the relationship between metabolic pathway and molecular pathway for synthesizing new molecules, for instance optimizing drug metabolization. In bioinformatics fields, multi-label prediction of metabolic pathways is a typical manner to understand this relationship. Graph neural networks (GNNs) have become an effective method to extract molecular structure's features for multi-label prediction of metabolic pathways. Though GNNs can effectively capture structural features from molecular structure graphs, building a well-performed GNN model for a given molecular structure data set requires the manual design of the GNN architecture and fine-tuning of the hyperparameters, which are time-consuming and rely on expert experience. To address the above challenge, we design an end-to-end automatic molecular structure representation learning framework named AutoMSR that can design the optimal GNN model based on a given molecular structure data set without manual intervention. We propose a multi-seed age evolution (MSAE) search algorithm to identify the optimal GNN architecture from the GNN architecture subspace. For a given molecular structure data set, AutoMSR first uses MSAE to search the GNN architecture, and then it adopts a tree-structured parzen estimator to obtain the best hyperparameters in the hyperparameters subspace. Finally, AutoMSR automatically constructs the optimal GNN model based on the best GNN architecture and hyperparameters to extract the molecular structure features for multi-label metabolic pathway prediction. We test the performance of AutoMSR on the real data set KEGG. The experiment results show that AutoMSR outperforms baseline methods on different multi-label classification evaluation metrics.

Side effects of drugs have gained increasing attention in the biomedical field, and accurate identification of drug side effects is essential for drug development and drug safety surveillance. Although the traditional pharmacological experiments can accurately detect the side effects of drugs, the identifying process is time-consuming, costly, and may lead to incomplete identification of side effects. With the expanding of various biomedical databases, many computational methods have been developed for the task of drug-side effect associations (DSAs) prediction. However, existing methods have the following three drawbacks: 1). multiple drug-related databases are not fully used; 2). the complex semantics among drugs and side effects are not effectively captured; 3). the explainability of the predicted DSAs is missed for most existing methods. Therefore, there is an urgent need to find a more effective method for predicting DSAs. To address these issues, we propose a novel meta-path-based graph neural network model for drug-side effect associations prediction (MPGNN-DSA). In MPGNN-DSA, a heterogeneous information network is first constructed by combining multiple biological datasets. Then, a meta-path-based feature learning module is utilized for learning high-quality representations of drugs and side effects by capturing the semantics contained in meta-paths of the constructed HIN. With the learned features, the prediction module is conducted to derive the predicted side effects for drugs. In addition, the explainability of the predicted DSAs can be provided as well with the semantics contained in meta-paths. We conduct comprehensive experiments, and the results demonstrate the effectiveness of MPGNN-DSA, suggesting that the proposed method will be a feasible solution to the task of DSAs prediction.

Glycans play an indispensable role in various bio-logical processes, such as cancer and autoimmune diseases. The function of glycan is closely determined by its structure. Due to the branch and nonlinear properties of glycans, previous research treats the glycans graph structure as a topological graph to represent glycans data effectively. Graph neural networks (GNNs) are an efficient graph mining method and have many applications in bioinformatics. Therefore, researchers have successfully used handcrafted GNNs to predict glycan immunogenicity. However, a GNN architecture contains many different components, and designing GNN architectures for specific graphs in the bioinformatics field is time-consuming and expert-dependent. To address this challenge, we propose an efficient automatic graph neural network method called EAGNN that can efficiently and automatically construct GNN architecture for glycan immunogenicity prediction. We design an effective graph attention pooling (GAP) search space. We use differential architecture search to efficiently create the optimal GNN architecture in the search space to build the GNN model for glycan immunogenicity prediction. We test EAGNN on the data set SugarBase based on the glycan immunogenicity prediction task. The experiment results show that EAGNN can work more superiorly than the baseline model and achieve the best performance.

Building a knowledge graph is a time-consuming and costly process which often applies complex natural language processing (NLP) methods for extracting knowledge graph triples from text corpora. Pre-trained large Language Models (PLM) have emerged as a crucial type of approach that provides readily available knowledge for a range of AI applications. However, it is unclear whether it is feasible to construct domain-specific knowledge graphs from PLMs. Motivated by the capacity of knowledge graphs to accelerate data-driven materials discovery, we explored a set of state-of-the-art pre-trained general-purpose and domain-specific language models to extract knowledge triples for metal-organic frameworks (MOFs). We created a knowledge graph benchmark with 7 relations for 1248 published MOF synonyms. Our experimental results showed that domain-specific PLMs consistently outperformed the general-purpose PLMs for predicting MOF related triples. The overall benchmarking results, however, show that using the present PLMs to create domain-specific knowledge graphs is still far from being practical, motivating the need to develop more capable and knowledgeable pre-trained language models for particular applications in materials science.

Despite recent progress in deep learning, visual emotion recognition remains a challenging problem due to ambiguity of emotion perception, diverse concepts related to visual emotion and lack of large-scale annotated dataset. In this paper, we present a large-scale multimodal pre-training method to learn visual emotion representation by aligning emotion, object, attribute triplet with a contrastive loss. We conduct our pre-training on a large web dataset with noisy tags and fine-tune on visual emotion classification datasets. Our method achieves state-of-the-art performance for visual emotion classification.

The increasing prevalence of antibiotic resistance has become a global health crisis. For the purpose of safety regulation, it is of high importance to identify antibiotic resistance genes (ARGs) in bacteria. Although culture-based methods can identify ARGs relatively more accurately, the identifying process is time-consuming and specialized knowledge is required. With the rapid development of whole genome sequencing technology, researchers attempt to identify ARGs by computing sequence similarity from public databases. However, these computational methods might fail to detect ARGs due to the low sequence identity to known ARGs. Moreover, existing methods cannot effectively address the issue of multidrug resistance prediction for ARGs, which is a great challenge to clinical treatments. To address the challenges, we propose an end-to-end multi-label learning framework for predicting ARGs. More specifically, the task of ARGs prediction is modeled as a problem of multi-label learning, and a deep neural network-based end-to-end framework is proposed, in which a specific loss function is introduced to employ the advantage of multi-label learning for ARGs prediction. In addition, a dual-view modeling mechanism is employed to make full use of the semantic associations among two views of ARGs, i.e. sequence-based information and structure-based information. Extensive experiments are conducted on publicly available data, and experimental results demonstrate the effectiveness of the proposed framework on the task of ARGs prediction.

Understanding the relationship between molecular structure and metabolic pathway classes is significant for optimizing drug metabolization. In bioinformatics, graph neural networks (GNNs) can effectively capture structural and semantic features for molecular representation. Graph neural networks have become an essential method to encode molecular structures for multi-label prediction of metabolic pathways. However, building a GNN model for a given molecular structure dataset requires the manual design of GNN structure and fine-tuning of the hyperparameters for the GNN model, which is time-consuming and relies on expert experience. In this paper, we design an automatic end-to-end molecular structure representation learning framework named Auto-MSR that can design a GNN model for molecular structure encoding with little manual intervention. For a given compound molecule structure dataset, Auto-MSR first uses an efficient pa rallel GNN structure search algorithm to identify the optimal GNN structure from the GNN structure subspace. Then, it adopts a tree-structured parzen estimator approach to obtain the best hyperparameters of the GNN model in the hyperparameters subspace. We test AutoMSR on the dataset KEGG based on the multi-label metabolic pathway prediction task. The comparing results show that AutoMSR outperforms state-of-art manual graph neural networks on performance.

Emotion understanding from an image is an important computer vision research topic, but the reasoning behind the justification has been largely unexplored. We propose to utilize language, and more specifically, explanation of the judgement to increase the explainability for this task. We collect a dataset with image, emotion tag and explanation in natural language, and conduct analysis on the dataset to gain insights on the ambiguity of the human emotion perceptual. We examine baseline methods to predict emotion from image, explanation and general image description, and unifying both modalities. Our experiments shed lights on effects from different modalities and we also identify opportunities for future visual emotion categorization research based on the analysis. We release our dataset to advance future research.

Scientific literature presents a wellspring of cutting-edge knowledge for materials science, including valuable data (e.g., numerical data from experiment results, material properties and structure). These data are critical for accelerating materials discovery by data-driven machine learning (ML) methods. The challenge is, it is impossible for humans to manually extract and retain this knowledge due to the extensive and growing volume of publications.To this end, we explore a fine-tuned BERT model for extracting knowledge. Our preliminary results show that our fine-tuned Bert model reaches an f-score of 85% for the materials named entity recognition task. The paper covers background, related work, methodology including tuning parameters, and our overall performance evaluation. Our discussion offers insights into our results, and points to directions for next steps.

In this position paper, we describe research on knowledge graph-empowered materials science prediction and discovery. The research consists of several key components including ontology mapping, materials data annotation, and information extraction from unstructured scholarly articles. We argue that although big data generated by simulations and experiments have motivated and accelerated the data-driven science, the distribution and heterogeneity of materials science-related big data hinders major advancements in the field. Knowledge graphs, as semantic hubs, integrate disparate data and provide a feasible solution to addressing this challenge. We design a knowledge-graph based approach for data discovery, extraction, and integration in materials science.

Recent years have witnessed the rapidly growing of the amount of semistructured documents in real-world applications. Due to the huge size of the real-world data, how to manage semistructured documents effectively is a big challenge for researchers. As a fundamental task in natural language processing field, document classification is a feasible way to handle the large-scale semistructured documents. However, existing methods fail to explicitly take advantage of the hierarchical semantics in semistructured documents. It's known that the contained semantics is beneficial for understanding the semistructured documents. Considering the hierarchical structure of a given semistructured document, we propose a semistructured document classification framework which explicitly utilizes the semantic hierarchical attention mechanism. More specifically, the hierarchical attention mechanism and graph neural network are employed to model semistructured documents, by which the multilevel semantic relationships and grammatical information are considered. Moreover, we propose an adaptive class cost learning method to treat the issue of data imbalance. Comprehensive experiments are conducted on two real-world data sets, and the results demonstrate that our framework performs better than selected baselines for semistructured document classification.

Research of Protein-Protein Interaction (PPI) Network Alignment is playing an important role in understanding the crucial underlying biological knowledge such as functionally homologous proteins and conserved evolutionary pathways across different species. Existing methods of PPI network alignment often try to improve the coverage ratio of the alignment result by aligning all proteins from different species. However, there is a fundamental biological premise that needs to be considered carefully: not every protein in a species can, nor should, find its homologous proteins in other species. In this work, we propose a novel alignment method to map only those proteins with the most similarity throughout the PPI networks of multiple species. For the similarity features of the protein in the networks, we integrate both topological features with biological characteristics to provide enhanced supports for the alignment procedures. For topological features, we apply a representation learning method on the networks and generate a low dimensional vector embedding with its surrounding structural features for each protein. The topological similarity of proteins from different PPI networks can thus be transferred as the similarity of their corresponding vector representations, which provides a new way to comprehensively quantify the topological similarities between proteins. We also propose a new measure for the topological evaluation of the alignment results which better uncover the structural quality of the alignment across multiple networks. Both biological and topological evaluations on the alignment results of real datasets demonstrate our approach is promising and preferable against previous multiple alignment methods.

Reconstructing gene networks from gene expression data is a long-standing challenge. In most applications, the observations can be divided into several distinct but related subpopulations and the gene expression measurements can be collected from multiple data types. Most existing methods are designed to estimate a single gene network from a single dataset. These methods may be suboptimal since they do not exploit the similarities and differences among different subpopulations and data types. In this article, we propose a joint graphical model to estimate the multiple gene networks simultaneously. Our model decomposes each subpopulation-specific gene network as a sum of common and unique components and imposes a group lasso penalty on gene networks corresponding to different data types. The gene network variations across subpopulations can be learned automatically by the decompositions of networks, and the similarities and differences among data types can be captured by the group lasso penalty. The simulation studies demonstrate that our method outperforms the state-of-the-art methods. We also apply our method to the cancer genome atlas breast cancer datasets to reconstruct subtype-specific gene networks. Hub nodes in the estimated subnetworks unique to individual cancer subtypes rediscover well-known genes associated with breast cancer subtypes and provide interesting predictions.

To identify and explore the similarities between diseases is of great significance for u nderstanding t he pathogenic mechanisms of emerging complex diseases. Some methods try to measure the similarity of diseases through deep learning models. However, the insufficient number of labelled similar disease pairs cannot support the optimal training of the models. In this paper, we propose a Multi-Task Graph Neural Network (MTGNN) framework to retrieve similar diseases by few-shot learning. To deal with the problem of insufficient number o f labelled similar disease pairs, we design double tasks to optimize the graph neural network for disease similarity task (lack of labelled training data) by introducing link prediction task (sufficient labelled training data). The similarity between diseases can then be obtained by measuring the distance between disease embeddings in high-dimensional space learning from the double tasks. The experiment results illustrate the overall effectiveness by comparing with prior methods on few labeled training dataset.

Many datasets that exists in the real world are often comprised of different representations or views which provide complementary information to each other. To integrate information from multiple views, data integration approaches such as nonnegative matrix factorization (NMF) have been developed to combine multiple heterogeneous data simultaneously to obtain a comprehensive representation. In this paper, we proposed a novel variant of symmetric nonnegative matrix factorization (SNMF), called Laplacian regularization based joint symmetric nonnegative matrix factorization (LJ-SNMF) for clustering multi-view data. We conduct extensive experiments on several realistic datasets including Human Microbiome Project data. The experimental results show that the proposed method outperforms other variants of NMF, which suggests the potential application of LJ-SNMF in clustering multi-view datasets. Additionally, we also demonstrate the capability of LJ-SNMF in community finding.

The development and prognosis of complex diseases usually involves changes in regulatory relationships among biomolecules. Understanding how the regulatory relationships change with genetic alterations can help to reveal the underlying biological mechanisms for complex diseases. Although several models have been proposed to estimate the differential network between two different states, they are not suitable to deal with situations where the molecules of interest are affected by other covariates. Nor can they make use of prior information that provides insights about the structures of biomolecular networks. In this study, we introduce a novel weighted fused conditional Gaussian graphical model to jointly estimate two state-specific biomolecular regulatory networks and their difference between two different states. Unlike previous differential network estimation methods, our model can take into account the related covariates and the prior network information when inferring differential networks. The effectiveness of our proposed model is first evaluated based on simulation studies. Experiment results demonstrate that our model outperforms other state-of-the-art differential networks estimation models in all cases. We then apply our model to identify the differential gene network between two subtypes of glioblastoma based on gene expression and miRNA expression data. Our model is able to discover known mechanisms of glioblastoma and provide interesting predictions.

Identifying gene network rewiring under different biological conditions is important for understanding the mechanisms underlying complex diseases. Gaussian graphical models, which assume the data follow the multivariate normal distribution, are widely used to identify gene network rewiring. However, the normality assume often fails in reality since the data are contaminated by extreme outliers in general. In this study, we propose a new robust differential graphical model to identify gene network rewiring between two conditions based on the multivariate t-distribution. The multivariate t-distribution is more robust to outliers than the normal distribution since it has heavy tails and allows values far from the mean. A fused lasso penalty is used to borrow information across conditions to improve the results. We develop an expectation maximization algorithm to solve the optimization model. Experiment results on simulated data show that our method outperforms the state-of-the-art methods. Our method is also applied to identify gene network rewiring between luminal A and basal-like subtypes of breast cancer, and gene network rewiring between the proneural and mesenchymal subtypes of glioblastoma. Several key genes which drive gene network rewiring are discovered.

Liver segmentation is a critical step in diagnosing various kinds of hepatic diseases. Based on the segmentation results, physicians can make further assessments more accurately. Although deep learning methods have achieved excellent performance in liver segmentation tasks, the traditional convolution encoder-decoder architecture may easily loss the spatial information due to the stacked convolution and pooling layers. In this paper, we present a non-local spatial feature based neural network (referred as NL-Net) to learn more spatial features of liver for more accurate segmentation. The NL-Net consists of an encoder block, a non-local spatial feature learning block and a decoder block. We utilized the pretrained ResNet model with transfer learning as the encoder. The non-local block can learn long range dependencies of the liver pixel position by computing the response at a position as a weighted sum of the responses at all positions, which can help the network learn more robust features. We applied the proposed model to ISBI 2019 CHAOs liver Segmentation Challenge task and evaluated it on the testing set. Experimental results show that the proposed NL-Net achieved an average dice of 0.972, RAVD of 1.593, ASSD of 1.926 and MSSD of 110.658 on the segmentation results.

Quantifying the similarities between diseases is now playing an important role in biology and medicine, which provides reliable reference information in finding similar diseases. Most of the previous methods for similarity calculation between diseases either use a single-source data or do not fully utilize multi-sources data. In this study, we propose an approach to measure disease similarity by utilizing multiple heterogeneous disease information networks. Firstly, multiple disease-related data sources are formulated as heterogeneous disease information networks which include various types of objects such as disease, pathway, and chemicals. Then, the corresponding subgraphs of these heterogeneous disease information networks are obtained by filtering vertices. Topological scores and semantics scores are calculated in these heterogenous subgraphs using Dynamic Time Warping (DTW) algorithm and meta path method respectively. In this way, we transform multiple heterogeneous disease networks to a homogeneous disease network with different weights on the edges. Finally, the disease nodes can be embedded according to the weights and the similarity between diseases can then be calculated using these n-dimensional vectors. Experiments based on benchmark set fully demonstrate the effectiveness of our method in measuring the similarity of diseases through multi-sources data.

Deep neural networks trained by medical images with dense annotations have revealed favourable performance on accurate organ segmentation. The current supervised methods demand voxel-level annotations which are not easily accessible due to the consuming of time and requirements of specialized knowledge and skills. In this paper, we propose a weakly supervised method based on a recurrent residual convolutional neural network trained only with image-level labels to generate voxel-level segmentation. The recurrent residual convolutional units take advantage of contextual information of successive slices and a spatial pooling layer is introduced after the last convolutional layer to aggregate local features and learn accurate localization. The final segmentation mask is computed by applying a conditional random field for spatial prediction. Our method shows competitive performance to fully supervised methods on the public NIH-CT-82 dataset for pancreas segmentation.

Accumulating evidence has demonstrated that RNAs play an important role in identifying various complex human diseases. However, the number of known disease related RNAs is still small and many biological experiments are time-consuming and labor-intensive. Therefore, researchers have focused on developing useful computational algorithms to predict associations between diseases and RNAs. It is useful for people to identify complex human diseases at molecular level, especially in diseases diagnosis, therapy, prognosis and monitoring. In this paper, we propose a novel framework Graph Convolutional Attention Network(GCAN) to predict potential disease-RNAs associations. Facing thousands of associations, GCAN benefits from the efficiency of deep learning model. Compared to other disease-RNAs association prediction methods, GCAN operates the computation process from global structure of disease-RNAs network with graph convolution networks(GCN) and can also integrate local neighborhoods with the attention mechanism. What is more, GCAN is at the first attempt to utilize GCN to discover the feature representation of the latent nodes in disease-RNAs network. In order to evaluate the performance of GCAN, we conduct experiments on two different disease-RNAs networks: disease-miRNA and disease-lncRNA. Comparisons of several state-of-the-art methods using disease-RNAs networks show that our novel frameworks outperform baselines by a wide margin in potential disease-RNAs associations.

Biomedical document classification is a fundamental task in biomedical field. Existing methods do not make full use of the hierarchically semantic structures in biomedical documents which can be utilized to improve the performance of biomedical document classification. In this paper, according to the hierarchical structures in given biomedical documents, we propose two models for biomedical document classification, which are based on the semantically hierarchical attention mechanism. Specifically, we utilize a hierarchical attention mechanism to model biomedical documents, taking into account simultaneously multiple-level semantic relationships in documents. In addition, an adaptive cost sensitive learning method is proposed to address the data imbalance issue. Extensive experiments on two real-world datasets demonstrate the effectiveness of the proposed methods.

Opinion mining has raised growing interest both in industry and academia in the past decade. Opinion role labeling (ORL) is a task to extract opinion holder and target from natural language to answer the question “who express what”. Recent years, neural network based methods with additional lexical and syntactic features have achieved state-of-the-art performances in similar tasks. Moreover, Bidirectional Encoder Representations from Transformers (BERT) has shown impressive performances among a variety of natural language processing (NLP) tasks. To investigate BERT based end-to-end model in ORL, we propose models using BERT, Bidirectional Long short-term Memory (BiLSTM) and Conditional Random Field (CRF) to jointly extract opinion roles (e.g., opinion holder and target). Experimental results show that our models achieve remarkable scores without using extra syntactic and/or semantic features. To our best knowledge, we are among the pioneers to successfully integrate BERT in this manner. Our work contributes to the improvement of state-of-the-art aspect-level opinion mining methods and providing strong baselines for future work.

SUMMARY: To identify biological network rewiring under different conditions, we develop a user-friendly R package, named DiffNetFDR, to implement two methods developed for testing the difference in different Gaussian graphical models. Compared to existing tools, our methods have the following features: (i) they are based on Gaussian graphical models which can capture the changes of conditional dependencies; (ii) they determine the tuning parameters in a data-driven manner; (iii) they take a multiple testing procedure to control the overall false discovery rate; and (iv) our approach defines the differential network based on partial correlation coefficients so that the spurious differential edges caused by the variants of conditional variances can be excluded. We also develop a Shiny application to provide easier analysis and visualization. Simulation studies are conducted to evaluate the performance of our methods. We also apply our methods to two real gene expression datasets. The effectiveness of our methods is validated by the biological significance of the identified differential networks. AVAILABILITY AND IMPLEMENTATION: R package and Shiny app are available at https://github.com/Zhangxf-ccnu/DiffNetFDR. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: Imputation of dropout events that may mislead downstream analyses is a key step in analyzing single-cell RNA-sequencing (scRNA-seq) data. We develop EnImpute, an R package that introduces an ensemble learning method for imputing dropout events in scRNA-seq data. EnImpute combines the results obtained from multiple imputation methods to generate a more accurate result. A Shiny application is developed to provide easier implementation and visualization. Experiment results show that EnImpute outperforms the individual state-of-the-art methods in almost all situations. EnImpute is useful for correcting the noisy scRNA-seq data before performing downstream analysis. AVAILABILITY AND IMPLEMENTATION: The R package and Shiny application are available through Github at https://github.com/Zhangxf-ccnu/EnImpute. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: Malignant liver tumor is one of the main causes of human death. In order to help physician better diagnose and make personalized treatment schemes, in clinical practice, it is often necessary to segment and visualize the liver tumor from abdominal computed tomography images. Due to the large number of slices in computed tomography sequence, developing an automatic and reliable segmentation method is very favored by physicians. However, because of the noise existed in the scan sequence and the similar pixel intensity of liver tumors with their surrounding tissues, besides, the size, position and shape of tumors also vary from one patient to another, automatic liver tumor segmentation is still a difficult task. RESULTS: We perform the proposed algorithm to the Liver Tumor Segmentation Challenge dataset and evaluate the segmentation results. Experimental results reveal that the proposed method achieved an average Dice score of 68.4% for tumor segmentation by using the designed network, and ASD, MSD, VOE and RVD improved from 27.8 to 21, 147 to 124, 0.52 to 0.46 and 0.69 to 0.73, respectively after performing adversarial training strategy, which proved the effectiveness of the proposed method. CONCLUSIONS: The testing results show that the proposed method achieves improved performance, which corroborated the adversarial training based strategy can achieve more accurate and robustness results on liver tumor segmentation task.

Gene dependency networks often undergo changes with respect to different disease states. Understanding how these networks rewire between two different disease states is an important task in genomic research. Although many computational methods have been proposed to undertake this task via differential network analysis, most of them are designed for a predefined data type. With the development of the high throughput technologies, gene activity measurements can be collected from different aspects (e.g., mRNA expression and DNA mutation). These different data types might share some common characteristics and include certain unique properties of data type. New methods are needed to explore the similarity and difference between differential networks estimated from different data types. In this study, we develop a new differential network inference model which identifies gene network rewiring by combining gene expression and gene mutation data. Similarities and differences between different data types are learned via a group bridge penalty function. Simulation studies have demonstrated that our method consistently outperforms the competing methods. We also apply our method to identify gene network rewiring associated with ovarian cancer platinum resistance from The Cancer Genome Atlas data. There are certain differential edges common to both data types and some differential edges unique to individual data types. Hub genes in the differential networks inferred by our method play important roles in ovarian cancer drug resistance.

Correlated anomaly detection (CAD) from streaming data is a type of group anomaly detection and an essential task in useful real-time data mining applications like botnet detection, financial event detection, industrial process monitor, etc. The primary approach for this type of detection in previous researches is based on principal score (PS) of divided batches or sliding windows by computing top eigenvalues of the correlation matrix, e.g. the Lanczos algorithm. However, this paper brings up the phenomenon of principal score degeneration for large data set, and then mathematically and practically prove current PS-based methods are likely to fail for CAD on large-scale streaming data even if the number of correlated anomalies grows with the data size at a reasonable rate; in reality, anomalies tend to be the minority of the data, and this issue can be more serious. We propose a framework with two novel randomized algorithms rPS and gPS for better detection of correlated anomalies from large streaming data of various correlation strength. The experiment shows high and balanced recall and estimated accuracy of our framework for anomaly detection from a large server log data set and a U.S. stock daily price data set in comparison to direct principal score evaluation and some other recent group anomaly detection algorithms. Moreover, our techniques significantly improve the computation efficiency and scalability for principal score calculation.

Aspect-level opinion mining aims to find and aggregate opinions on opinion targets. Previous work has demonstrated that precise modeling of opinion targets within the surrounding context can improve performances. However, how to effectively and efficiently learn hidden word semantics and better represent targets and the context still needs to be further studied. In this paper, we propose and compare two interactive attention neural networks for aspect-level opinion mining, one employs two bi-directional Long-Short-Term-Memory (BLSTM) and the other employs two Convolutional Neural Networks (CNN). Both frameworks learn opinion targets and the context respectively, followed by an attention mechanism that integrates hidden states learned from both the targets and context. We compare our model with state-of-the-art baselines on two SemEval 2014 datasets1. Experiment results show that our models obtain competitive performances against the baselines on both datasets. Our work contributes to the improvement of state-of-the-art aspect-level opinion mining methods and offers a new approach to support human decision-making process based on opinion mining results. The quantitative and qualitative comparisons in our work aim to give basic guidance for neural network selection in similar tasks.

Summary: We develop DiffGraph, an R package that integrates four influential differential graphical models for identifying gene network rewiring under two different conditions from gene expression data. The input and output of different models are packaged in the same format, making it convenient for users to compare different models using a wide range of datasets and carry out follow-up analysis. Furthermore, the inferred differential networks can be visualized both non-interactively and interactively. The package is useful for identifying gene network rewiring from input datasets, comparing the predictions of different methods and visualizing the results. Availability and implementation: The package is available at https://github.com/Zhangxf-ccnu/DiffGraph. Contact: [email protected]. Supplementary information: Supplementary data are available at Bioinformatics online.

BACKGROUND: In bioinformatics, network alignment algorithms have been applied to protein-protein interaction (PPI) networks to discover evolutionary conserved substructures at the system level. However, most previous methods aim to maximize the similarity of aligned proteins in pairwise networks, while concerning little about the feature of connectivity in these substructures, such as the protein complexes. RESULTS: In this paper, we identify the problem of finding conserved protein complexes, which requires the aligned proteins in a PPI network to form a connected subnetwork. By taking the feature of connectivity into consideration, we propose ConnectedAlign, an efficient method to find conserved protein complexes from multiple PPI networks. The proposed method improves the coverage significantly without compromising of the consistency in the aligned results. In this way, the knowledge of protein complexes in well-studied species can be extended to that of poor-studied species. CONCLUSIONS: We conducted extensive experiments on real PPI networks of four species, including human, yeast, fruit fly and worm. The experimental results demonstrate dominant benefits of the proposed method in finding protein complexes across multiple species.

BACKGROUND: The traditional methods of visualizing high-dimensional data objects in low-dimensional metric spaces are subject to the basic limitations of metric space. These limitations result in multidimensional scaling that fails to faithfully represent non-metric similarity data. RESULTS: Multiple maps t-SNE (mm-tSNE) has drawn much attention due to the construction of multiple mappings in low-dimensional space to visualize the non-metric pairwise similarity to eliminate the limitations of a single metric map. mm-tSNE regularization combines the intrinsic geometry between data points in a high-dimensional space. The weight of data points on each map is used as the regularization parameter of the manifold, so the weights of similar data points on the same map are also as close as possible. However, these methods use standard momentum methods to calculate parameters of gradient at each iteration, which may lead to erroneous gradient search directions so that the target loss function fails to achieve a better local minimum. In this article, we use a Nesterov momentum method to learn the target loss function and correct each gradient update by looking back at the previous gradient in the candidate search direction. By using indirect second-order information, the algorithm obtains faster convergence than the original algorithm. To further evaluate our approach from a comparative perspective, we conducted experiments on several datasets including social network data, phenotype similarity data, and microbiomic data. CONCLUSIONS: The experimental results show that the proposed method achieves better results than several versions of mm-tSNE based on three evaluation indicators including the neighborhood preservation ratio (NPR), error rate and time complexity.

Exploring the rewiring pattern of gene regulatory networks between different pathological states is an important task in bioinformatics. Although a number of computational approaches have been developed to infer differential networks from high-throughput data, most of them only focus on gene expression data. The valuable static gene regulatory network data accumulated in recent biomedical researches are neglected. In this study, we propose a new Gaussian graphical model-based method to infer differential networks by integrating gene expression and static gene regulatory network data. We first evaluate the empirical performance of our method by comparing with the state-of-the-art methods using simulation data. We also apply our method to The Cancer Genome Atlas data to identify gene network rewiring between ovarian cancers with different platinum responses, and rewiring between breast cancers of luminal A subtype and basal-like subtype. Hub genes in the estimated differential networks rediscover known genes associated with platinum resistance in ovarian cancer and signatures of the breast cancer intrinsic subtypes.

Observational data resources based on the capture of clinical data in the electronic health record (EHR) have produced significant learning opportunities in many areas of medicine. These large data resources can span multiple hospital systems and employ common semantics, ontologies, and data models. They have uncovered critical safety issues for patients, and spurred observational research and clinical decision support. In the age of precision medicine there is also an increased need to obtain genomic and clinical data to discover novel treatments for the deadliest of diseases. With this, there are efforts to create deep-dive disease specific repositories that include tissue in biobanks. The latter require significant human annotation of biospecimens. Securing the data is especially critical in rare pediatric brain tumors. In the specific case of The Children's Brain Tumor Tissue Consortium (CBTTC) an international rare pediatric brain tumor repository, the number of patients that need to be followed prospectively is outpacing the ability of human annotation. In this preliminary study, we perform a prescribed data exploration analysis on simulation data in the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM) employed by the pediatric data network PEDSNet with the intention to ascertain feasibility in automatic annotation of patient records in the CBTTC.

The human oral cavity is an important habitat of microbes in the human body. It includes the colonization of various microorganisms such as bacteria, archaea, fungi, protozoa and viruses. Although oral diseases have been studied for decades, we have limited understanding of the boundaries of a healthy oral ecosystem and ecological shift toward dysbiosis. Here, we analyzed salivary microbiomes from 268 healthy adults after overnight fasting. The microbiome data set is firstly divided into five sample clusters based on the similarity pattern of microbial abundance. For each cluster, the correlation networks among salivary bacteria are constructed based on an ensemble of six correlations and two dissimilarity measure. The stability and specificity of modularity in the five microbial networks are investigated. The existences of conserved and changing modules were found across five microbial correlation networks.

From a biological standpoint, due to the special combination of complex symptoms, some type of complex diseases is difficult to be accurately diagnosed. Known as phenotypic overlap, these sets of disease-related symptoms reveal a common pathological and physiological mechanism. Researchers attempt to visualize the phenotypic relationships between different human diseases from the perspective of machine learning, but traditional methods of visualizing high-dimensional data objects into low-dimensional would be subject to fundamental limitations of metric spaces. Our method is primarily based on the multiple maps t-SNE regularization, which is a probabilistic method for visualizing data points in multiple low-dimensional spaces. We use the Nesterov accelerated gradient method to learn the objective loss function. This method thought to counterweigh too high velocities by “peeking ahead” actual objective values in the candidate search direction, thus providing a larger and timelier correction to velocity. Experiments results on several dataset show that the proposed method outperforms the original version of mm-tSNE and mm-tSNE with regularization, as measured by the neighborhood preservation ratio. This suggests the modified mm-tSNE regularization can be applied directly in other domain including social and biological datasets.

There are intensive computational efforts to discover large-scale microbial interactions from metagenomic abundance data, however, it is often difficult to validate such inferred interactions without a manually curated dataset. There are also a number of small-scale microbial interactions reported in massive literature with experimental confidence. Text mining can be employed to extract such microbial interactions from biomedical literature which could be a significant complement to abundance-based method. The key tasks of text mining include named entity recognition and relation extraction. Named entity recognition identifies the name of the specified type from the text. We manually annotated a corpus with 1344 abstracts from microbial literature for the task of bacterial named entity recognition. Six new features were added in addition to the general features of the biomedical field. Based on a bacterial dictionary and conditional random field (CRF), the bacterial named entity recognition model was trained and it achieved a performance with precision 89.118%, recall 81.598 % and F-measure 85.192%. The system and template are available at https://github.com/bluelilywxy/BacNER-V1.0.git.

Gene (microRNA) identification is a key step in understanding the cellular mechanisms. Compared with biological experiments, computational prediction of disease genes is cheaper and more effortless. In this study, we analyzed the properties of tumor-associated microRNA in mouse and found that tumor-associated genes display 8distinguishingfeatures when compared with genes not yet known to be involved in tumor. The features of tumor-associated genes tend to located at network center and interact with each other were found by analyze the network characteristics. In addition, the features of the tumor-associated genes tend to be involved in certain biological processes and show certain phenotypes also were found through enrichment analysis. Based on these features, a machine-learning algorithm SVM were developed to predict new tumor-associated genes in mouse. Using the machine-learning algorithm, 120 tumor-associated genes were predicted with a posterior probability more than 0.9. We verified the accuracy of the identification framework with the data set of tumor-associated genes, and the result shows that this method is feasible.

Genome-wide association studies (GWAS) of T2D have discovered a number of loci that contribute to susceptibility to the disease. In this paper, we classified and identified the suspected risky Loci of T2D with computational method based on the known T2D GWAS-associated SNPs. The framework includes two parts: we first classified the SNPs based on their features of position and function through a simplified classification decision tree which was constructed by C4.5 decision tree algorithm; we then identified whether the genes associated with the suspected risky SNPs are associated with T2D by using random walk algorithm with Restart Model on the PPI network of T2D GWAS-associated genes among proteins and interactors. Based on the classification of SNP associated with T2D, we analyzed molecular pathogenesis of T2D. We verified the accuracy and reliability of the classification and identification framework with the data set of GWAS-associated SNPs. The result shows that this method is reliable. It provides a significant way to identify and classify the suspected risky Loci associated with T2D and further insights into the molecular pathogenesis of T2D.

In this paper, we computationally predicted the interactions between HIV-1 and human proteins, based on the hypothesis that proteins with similar interface architecture share similar interaction partners. Evolution - aware protein structural alignment method UniAlign was used to calculate the similarity between two protein interface architectures. Using experimentally verified HIV-1, human protein-protein interactions data, we first selected 12 features, including geometric similarity, conversion similarity etc.; then trained a support vector machine (SVM) with Gaussian kernel for the binary classification problem: whether a given protein pairs `interact' or `no interact'. We used the trained and tuned SVM classifier to discover potential novel HIV-1 interacting partners for human proteins. Many predicted interactions had significant literature support, and we modeled the novel 3D interacting complex for HIV-1 envelope gp120 and gp41 proteins. We provided the first structural evidence for those interactions.

Botnet, a group of coordinated bots, is becoming the main platform of malicious Internet activities like DDOS, click fraud, web scraping, spam/rumor distribution, etc. This paper focuses on design and experiment of a new approach for botnet detection from streaming web server logs, motivated by its wide applicability, real-time protection capability, ease of use and better security of sensitive data. Our algorithm is inspired by a Principal Component Analysis (PCA) to capture correlation in data, and we are first to recognize and adapt Lanczos method to improve the time complexity of PCA-based botnet detection from cubic to sub-cubic, which enables us to more accurately and sensitively detect botnets with sliding time windows rather than fixed time windows. We contribute a generalized online correlation matrix update formula, and a new termination condition for Lanczos iteration for our purpose based on error bound and non-decreasing eigenvalues of symmetric matrices. On our dataset of an ecommerce website logs, experiments show the time cost of Lanczos method with different time windows are consistently only 20% to 25% of PCA.

Known as phenotypic overlapping, some disease-rel ated symptoms share a common pathologi cal and physiological mechanism. Researchers attempt to visualize the phenotypic relationships between different human diseases from the perspective of machine learning, but traditional visualization methods may be subject to fundamental limitations of metric spaces. Multiple maps t-SNE regularization method, a probabilistic method for visualizing data points in multiple low-dimensional spaces has been proposed to address the limitation. However, the convergence speed is low when apply on the scale dataset. We use the RMSProp with Nesterov momentum method to learn the objective loss function. This method normalize the gradients by applying an exponential moving average of gradient magnitude for each iteration parameter and use Nesterov momentum to counterweigh too high velocities by “peeking ahead” actual objective values in the candidate search direction. This method convergent faster than the original method of convergence speed. Experiments results on several dataset shows that the proposed method outperforms the several version of mm-tSNE with or without regularization, as measured by the neighborhood preservation ratio and error rate. This suggests the modified mm-tSNE regularization can be applied directly in other domain including social, biological and microbiomic datasets.

This paper presents a non-trivial reconstruction of a previous joint topic-sentiment-preference review model TSPRA with stick-breaking representation under the framework of variational inference (VI) and stochastic variational inference (SVI). TSPRA is a Gibbs Sampling based model that solves topics, word sentiments and user preferences altogether and has been shown to achieve good performance, but for large dataset it can only learn from a relatively small sample. We develop the variational models vTSPRA and svTSPRA to improve the time use, and our new approach is capable of processing millions of reviews. We rebuild the generative process, improve the rating regression, solve and present the coordinate-ascent updates of variational parameters, and show the time complexity of each iteration is theoretically linear to the corpus size, and the experiments on Amazon datasets show it converges faster than TSPRA and attains better results given the same amount of time. In addition, we tune svTSPRA into an online algorithm ovTSPRA that can monitor oscillations of sentiment and preference overtime. Some interesting fluctuations are captured and possible explanations are provided. The results give strong visual evidence that user preference is better treated as an independent factor from sentiment.

Cross-Domain Collaborative Filtering(CDCF) methods transfer knowledge from auxiliary domains (e.g., books) to improve recommendation in a target domain (e.g. movies). Most CDCF methods exploit homogeneous user feedback, e.g. numeric ratings, from auxiliary domains as the knowledge source. However, in a typical recommender system, the usage data is usually heterogeneous and therefore is potential to better improve recommendation in other domains. In this paper, we propose a novel and generic CDCF solution called Heterogeneous Knowledge Transfer via Domain Regularization (HKT-DR). Our solution is able to mine high quality knowledge from heterogeneous knowledge sources, i.e. both explicit and implicit feedbacks from multiple auxiliary domains, by building a fused user similarity network, and to incorporate the knowledge by imposing domain regularization to constrain matrix factorization objective function. Extensive experiments on real world datasets show that the proposed HKT-DR model outperforms the state-of-the-art CDCF solutions.

Big data in education relate closely to a wealth of activities from the teachers, students and parents, as well as a substantial resources of knowledge that can be represented in hierarchical structures. The activities have characteristic of geolocation which can be projected onto a map, while the resources of knowledge can also be converted into the map. A map-based management and visual analysis method will largely benefit the users and the researchers from taking advantages of the big data in education. In this paper, we propose a novel map based method to manage and analyze the big data of mobile learning in education. With this method, the activities of users scattered among the space are reorganized on a geographic map with location changes in time series, and the resources are geo-tagged with the information from the developers or adopters, which are converted to a map style according to their hierarchical structures even when the users' information are unavailable. We first present the basic framework to organize the data by a map-based technology, and then a platform is proposed to perform the visual analysis. The method is adapt to the construction of massive online learning system and the mobile learning system of Central China Normal University to serve a national wide big data cloud learning program.

Network alignment is becoming an active topic in network data analysis. Despite extensive research, we realize that efficient use of topological and attribute information for large attributed network alignment has not been sufficiently addressed in previous studies. In this paper, based on Stochastic Block Model (SBM) and Dirichlet-multinomial, we propose "divide-and-conquer" models CAlign that jointly consider network alignment, community discovery and community alignment in one framework for large networks with node attributes, in an effort to reduce both the computation time and memory usage while achieving better or competitive performance. It is provable that the algorithms derived from our model have sub-quadratic time complexity and linear space complexity on a network with small densification power, which is true for most real-world networks. Experiments show CAlign is superior to two recent state-of-art models in terms of accuracy, time and memory on large networks, and CAlign is capable of handling millions of nodes on a modern desktop machine.

Word embedding in the NLP area has attracted increasing attention in recent years. The continuous bag-of-words model (CBOW) and the continuous Skip-gram model (Skip-gram) have been developed to learn distributed representations of words from a large amount of unlabeled text data. In this paper, we explore the idea of integrating extra knowledge to the CBOW and Skip-gram models and applying the new models to biomedical NLP tasks. The main idea is to construct a weighted graph from knowledge bases (KBs) to represent structured relationships among words/concepts. In particular, we propose a GCBOW model and a GSkip-gram model respectively by integrating such a graph into the original CBOW model and Skip-gram model via graph regularization. Our experiments on four general domain standard datasets show encouraging improvements with the new models. Further evaluations on two biomedical NLP tasks (biomedical similarity/relatedness task and biomedical Information Retrieval (IR) task) show that our methods have better performance than baselines.

In this paper, we propose a novel k-anonymization scheme to counter deanonymization queries on social networks. With this scheme, all entities are protected by k-anonymization, which means the attackers cannot re-identify a target with confidence higher than 1/k. The proposed scheme minimizes the modification on original networks, and accordingly maximizes the utility preservation of published data while achieving k-anonymization privacy protection. Extensive experiments on real data sets demonstrate the effectiveness of the proposed scheme, where the efficacy of the k-anonymized networks is verified with the distributions of pagerank, betweenness, and their Kolmogorov-Smirnov (K-S) test.

BACKGROUND: Essential proteins are indispensable to the survival and development process of living organisms. To understand the functional mechanisms of essential proteins, which can be applied to the analysis of disease and design of drugs, it is important to identify essential proteins from a set of proteins first. As traditional experimental methods designed to test out essential proteins are usually expensive and laborious, computational methods, which utilize biological and topological features of proteins, have attracted more attention in recent years. Protein-protein interaction networks, together with other biological data, have been explored to improve the performance of essential protein prediction. RESULTS: The proposed method SCP is evaluated on Saccharomyces cerevisiae datasets and compared with five other methods. The results show that our method SCP outperforms the other five methods in terms of accuracy of essential protein prediction. CONCLUSIONS: In this paper, we propose a novel algorithm named SCP, which combines the ranking by a modified PageRank algorithm based on subcellular compartments information, with the ranking by Pearson correlation coefficient (PCC) calculated from gene expression data. Experiments show that subcellular localization information is promising in boosting essential protein prediction.

Microbiome datasets are often comprised of different representations or views which provide complementary information to understand microbial communities, such as metabolic pathways, taxonomic assignments, and gene families. Data integration methods including approaches based on nonnegative matrix factorization (NMF) combine multi-view data to create a comprehensive view of a given microbiome study by integrating multi-view information. In this paper, we proposed a novel variant of NMF which called Laplacian regularized joint non-negative matrix factorization (LJ-NMF) for integrating functional and phylogenetic profiles from HMP. We compare the performance of this method to other variants of NMF. The experimental results indicate that the proposed method offers an efficient framework for microbiome data analysis.

Microbiome datasets are often comprised of different representations or views which provide complementary information, such as genes, functions, and taxonomic assignments. Integration of multi-view information for clustering microbiome samples could create a comprehensive view of a given microbiome study. Similarity network fusion (SNF) can efficiently integrate similarities built from each view of data into a unique network that represents the full spectrum of the underlying data. Based on this method, we develop a Robust Similarity Network Fusion (RSNF) approach which combines the strength of random forest and the advantage of SNF at data aggregation. The experimental results indicate the strength of the proposed strategy. The method substantially improves the clustering performance significantly comparing to several state-of-the-art methods in several datasets.

Part of the unique cultural heritage of China is the Chinese couplet.Given a sentence (namely an antecedent clause), people reply with another sentence (namely a subsequent clause) equal in length.Moreover, a special phenomenon is that corresponding characters from the same position in the two clauses match each other by following certain constraints on semantic and/or syntactic relatedness.Automatic couplet generation by computer is viewed as a difficult problem and has not been fully explored.In this paper, we formulate the task as a natural language generation problem using neural network structures.Given the issued antecedent clause, the system generates the subsequent clause via sequential language modeling.To satisfy special characteristics of couplets, we incorporate the attention mechanism and polishing schema into the encoding-decoding process.The couplet is generated incrementally and iteratively.A comprehensive evaluation, using perplexity and BLEU measurements as well as human judgments, has demonstrated the effectiveness of our proposed approach.

This paper proposes a new HDP based online review rating regression model named Topic-Sentiment-Preference Regression Analysis (TSPRA). TSPRA combines topics (i.e. product as-pects), word sentiment and user preference as regression factors, and is able to perform topic clus-tering, review rating prediction, sentiment analysis and what we invent as “critical aspect” analysis altogether in one framework. TSPRA extends sentiment approaches by integrating the key concept “user preference” in collaborative filtering (CF) models into consideration, while it is distinct from current CF models by decoupling “user preference” and “sentiment” as independent factors. Our experiments conducted on 22 Amazon datasets show overwhelming better performance in rating predication against a state-of-art model FLAME (2015) in terms of error, Pearson’s Correlation and number of inverted pairs. For sentiment analysis, we compare the derived word sentiments against a public sentiment resource SenticNet3 and our sentiment estimations clearly make more sense in the context of online reviews. Last, as a result of the de-correlation of “user preference” from “sentiment”, TSPRA is able to evaluate a new concept “critical aspects”, defined as the prod-uct aspects seriously concerned by users but negatively commented in reviews. Improvement to such “critical aspects” could be most effective to enhance user experience.

Essential proteins play a crucial role in the survival and development process of life, as they provide all available nutrients to maintain life. Therefore, many researchers pay attention to the identification of essential proteins. As experiments methods are usually costly and time-consuming, more and more computational algorithms have been developed to discover essential proteins based on biological and topological features. Given that the subcellular localization is very important in understanding protein-protein interaction, in this paper, a novel method is proposed to predict essential proteins, which integrates the subcellular compartments information with Pearson correlation coefficient (PCC) of gene expression data. We name this method SCP in this paper. In order to evaluate the prediction performance of our method, several experiments are carried out to compare SCP with other methods. The results demonstrate that SCP has a better prediction performance of essential proteins than other methods.

Many datasets existed in the real world are often comprised of different representations or views which provide complementary information to each other. For example, microbiome datasets can be represented by metabolic paths, taxonomic assignment or gene families. To integrate information from multiple views, data integration approaches such as methods based on nonnegative matrix factorization (NMF) have been developed to combine multi-view information simultaneously to obtain a comprehensive view which reveals the underlying data structure shared by multiple views. In this paper, we proposed a novel variant of symmetric nonnegative matrix factorization (SNMF), called Laplacian regularized joint symmetric nonnegative matrix factorization (LJ-SNMF) for clustering multi-view data. We conduct extensive experiments on several realistic datasets including Human Microbiome Project (HMP) data. The experimental results show that the proposed method outperforms other variants of NMF, which suggests the potential application of LJ-SNMF in clustering multi-view datasets.

The microbiota living in the human body plays a very important role in our health and disease, so the identification of microbes associated with diseases will contribute to improving medical care and to better understanding of microbe functions, interactions. However, the known associations between the diseases and microbes are very less. We proposed a new method for prioritization of candidate microbes to predict disease-microbe relationships that based on the random walking on the heterogeneous network. Here, we first constructed a heterogeneous network by connecting the disease network and microbe network using the disease-microbe relationship information, then extended the random walk to the heterogeneous network, finally we used leave-one-out cross-validation to evaluate the method and ranked the candidate disease-causing microbes. We used the algorithm to disclose some potential association between disease and microbe that cannot be found by microbe network or disease network alone. Furthermore, we studied three representative diseases, Type 2 diabetes, Asthma and Psoriasis, and presented the potential microbes associated with these diseases, respectively. We confirmed that the discovery of the associations will be a good clinical solution for disease mechanism understanding, diagnosis and therapy.

Detection of protein complexes and functional modules plays a crucial role for strengthening the comprehension of cellular organization and biological functions on the dynamic protein-protein interaction network. In this article, we put forward a new strategy to identify temporal protein complexes. Integrating time-course gene expression data into static protein interaction data, a series of time-sequenced subnetworks were constructed. Then we combined the network topology and gene ontology information for defining the distance between proteins in PPI network. A novel method to find the cluster centers and then form initial clusters was based on the idea that cluster centers are usually recognized as nodes with higher densities than their neighbors and with a relatively larger distance from other cluster centers. Finally, inspired by the brainstorming discussion process, two ways are introduced to update the initial clusters for achieving the optimal results. After the filtering and merging procedure, experimental results demonstrated that the proposed strategy had a good performance comparing with the other four advanced algorithms - MCODE, FAG-EC, HC-PIN, and CNC.

Alignment among protein-protein interaction (PPI) networks largely benefits our understanding in biological researches as it contributes greatly to the uncovering of important information, such as evolutionary conserved pathways, protein complexes and functional orthologs. Global alignment of multiple PPI networks aims at clustering functionally conserved proteins throughout different species, where most traditional methods attempt to achieve results with high overall coverage and consistency. However little attention was paid on the deeper level criteria of k-coverage and especially the k-consistency that we additionally conducted for evaluations, where k indicates the number of species that the proteins in a cluster belong to. In this paper, we propose a novel approach for global alignment of multiple PPI networks which achieved high k-coverage and k-consistency simultaneously in addition to conventional criteria. The evaluations demonstrate the preferable comprehensive strength of our approach against other three state-of-the-art methods on six datasets composed of various numbers of PPI networks from five species.

Disease gene identification is a key step to understand the cellular mechanisms associated with a specific disease. Compared with biological experiments, computational predictions of disease genes are cheaper and more effortless. Many computational methods are used to detect causal genes for diseases on the protein-protein interaction (PPI) networks generated by the high-throughput technology. However, the accuracy of these methods need to be improved due to the false interactions in the PPI data. To deal with the challenge, other methods are proposed via the integration of biological information from different sources with the PPI networks. In this work, a new algorithm AIDG is developed to predict disease genes. First, the weighted PPI networks are built by incorporating the protein subcellular localization information into the human PPI networks. Next, all of disease candidate genes are scored in terms of a iteration function. Finally, they are ranked on descending order of their scores. The top candidates are considered as potential disease genes. The results from the leave-one-out crossing validation (LOOCV) show that AIDG outperforms other similar methods like DADA and ToppNet.

Welcome to the 2016 IEEE International Conference on Bioinformatics and Biomedicine (IEEE BIBM 2016) being held in the Shenzhen, China from December 15–18, 2016. On behalf of the IEEE BIBM 2016 Organizing Team, we would like to thank you for your participation and hope you enjoy the conference.

Single cell experimental studies provide an unprecedented opportunity to examine the heterogeneity of molecular processes in different cells. However, the reconstruction of a sequence of changes in molecular processes and development of regulatory networks using single cell data are still challenging problems in bioinformatics and systems biology. In this work we propose an integrated framework to infer genetic regulatory networks using single cell experimental data. We first use the Wanderlust algorithm to construct the pseudo-trajectory of gene expression activities. Due to noise in the expression data, a Gauss process regression method is employed to produce a smoothly trajectory. Our integrated approach includes both a top-down approach (i.e. the GENIE3 algorithm) to infer the network structure and a bottom-up approach (i.e. differential equation model) to reverse-engineering the regulatory network. Using the gene network of hematopoietic development in the mouse embryo as the test problem, we developed a dynamic model for a network of nine genes. Our results suggest that the proposed integrated framework is an effective approach to reconstruct regulatory networks from single cell data.

While much progress has been made on the genetic analysis of osteoporosis in the past 20 years, there are a lot of genes and SNPs that are associated with osteoporosis through GWAS. In this paper, we aim to identify the risky SNPs associated with osteoporosis by algorithms based on the known osteoporosis GWAS-associated SNPs. The whole framework of our prediction method includes two steps: Firstly, we identify whether the associated genes of the suspected risky SNPs is osteoporosis GWAS-associated genes by the method of random walk algorithm on the PPI network of osteoporosis GWAS-associated genes. Then, we classify the positive result SNPs based on their features of position and function through ID3 decision tree algorithm. We verify the accuracy of the prediction framework with the data set of GWAS-associated SNPs, and the result shows that the method is feasible. It provides a more convenient way to identify the risky SNPs of osteoporosis associated.

Understanding ongoing topics and their evolutions in social media is of great importance. Although topic analysis is not a novel research question, social media environment has presented new challenges. First, with insufficient co-occurrence information, short text have undermined many word co-occurrence oriented topic models' applicability. Second, real time message streams make traditional discretized topic tracking methods hard to function. Third, topics' evolution mechanisms are of great importance in social media context, but many studies have ignored them. Forth, topics have more complicated correlation among each other. Considering the existing problems, this paper has proposed a Semi-Supervised Dirichlet-Hawkes Process (SDHP) to deal with topic detection and tracking from social media. The main contributions of this paper are reflected in: (1) SDHP can handle short text problem efficiently; (2) SDHP can track topics from continuous message stream; (3) SDHP can reveal topics' underlying evolution patterns; and (4) SDHP can capture topics' correlations We have evaluated SDHP's ability in both topic detection and tracking in 8 real datasets from Twitter, and the algorithm's performances are very promising.

In the real world, many problems on massive graphs can be mapped to an underlying critical problem of discovering top-k subgraphs. For massive graphs, subgraph queries may have enormous number of matches, and so it is inefficient to compute all matches when only top-k matches are desired. Meanwhile, parallel algorithm is urgent for the scalability of massive graph computing. In this paper, we address the challenges of top-k subgraph query in massive graph. Firstly, we present a new graph matching notion: “approximate graph simulation”. With approximate graph simulation, top-k subgraph query can be customized by appointing a weighted query graph, which provides good flexibility for different application scenarios. Secondly, we propose a parallel top-k subgraph query algorithm at the level of vertex. With such algorithm, each vertex in massive graph obtains its matching state separately without requiring global graph information. In the algorithm, we also design a filter mechanism to speed up the the computation and a aggregation mechanism to obtain top-k vertices for query focus. Using real-life datasets, we experimentally verify that our approach of parallel top-k subgraph query are efficient.

Nowadays the explosion of Web information has led to the boom of massive web documents such as news webpages, online literature, etc. The latent topics behind the documents spread by self-evolution and mutual transition. Understanding how topics in documents evolve and transit is an important and challenging problem. Topic model is a set of powerful toolkits to model documents generation to find their underlying topics, usually at the unigram level, making it difficult to model the relationship between terms and their underlying topics. In this paper, we propose a pairwise topic modeling method to incorporate a pairwise relationship into topic modeling methods. We manage to discover latent topics as well as topic transitions at the same time in a natural way. We show that the pairwise topic model can facilitate discovering of individual topics as well as topic evolution. The results indicate our proposed method leads to a significant performance improvement over the traditional topic modeling methods, such as Latent Dirichlet Allocation (LDA) in terms of language perplexity. Besides, we conduct a series of empirical studies to show the topic words and topic transitions discovered. From the case studies, we show that with the help of PTM methods, people are able to explicitly understand how topics evolve and transit between each other.

Pattern mining is a fundamental topic in data mining area. Many pattern mining techniques, such as closed and maximal pattern mining have been proposed for different applications. However, when calculating the frequency of a pattern, the existing techniques treat each word equally. For example, although the word `pie' in `I love eating pie.' is quite different from `pie' in `american pie', `pie' in `american pie' will still be added up to the counts of `pie' when calculating its frequency. Therefore, this paper aims to overcome the drawback to find the valid patterns tailored to text mining. We will approach pattern mining from a different perspective and introduce a novel problem of frequent semantic pattern mining. We then propose an algorithm to solve this problem via suffix array sorting. The algorithm can be implemented to run in linear time. Compared with traditional pattern representations, our results show the semantic patterns extracted are more than 13% compact. Also, classifier built on these features is no less or more powerful.

In human-computer conversation systems, the context of a userissued utterance is particularly important because it provides useful background information of the conversation.However, it is unwise to track all previous utterances in the current session as not all of them are equally important.In this paper, we address the problem of session segmentation.We propose an embedding-enhanced TextTiling approach, inspired by the observation that conversation utterances are highly noisy, and that word embeddings provide a robust way of capturing semantics.Experimental results show that our approach achieves better performance than the TextTiling, MMD approaches.

Product reviews have become an important resource for customers before they make purchase decisions. However, the abundance of reviews makes it difficult for customers to digest them and make informed choices. In our study, we aim to help customers who want to quickly capture the main idea of a lengthy product review before they read the details. In contrast with existing work on review analysis and document summarization, we aim to retrieve a set of real-world user questions to summarize a review. In this way, users would know what questions a given review can address and they may further read the review only if they have similar questions about the product. Specifically, we design a two-stage approach which consists of question retrieval and question diversification. We first propose probabilistic retrieval models to locate candidate questions that are relevant to a review. We then design a set function to re-rank the questions with the goal of rewarding diversity in the final question set. The set function satisfies submodularity and monotonicity, which results in an efficient greedy algorithm of submodular optimization. Evaluation on product reviews from two categories shows that the proposed approach is effective for discovering meaningful questions that are representative for individual reviews.

In recent years, online social networks are among the most popular websites with high PV (Page View) all over the world, as they have renewed the way for information discovery and distribution. Millions of users have registered on these websites and hence generate formidable amount of user-generated contents every day. The social networks become "giants", likely eligible to carry on any research tasks. However, we have pointed out that these giants still suffer from their "Achilles Heel", i.e., extreme sparsity. Compared with the extremely large data over the whole collection, individual posting documents such as microblogs seem to be too sparse to make a difference under various research scenarios, while actually these postings are different. In this paper we propose to tackle the Achilles Heel of social networks by smoothing the language model via influence propagation. To further our previously proposed work to tackle the sparsity issue, we extend the socialized language model smoothing with bi-directional influence learned from propagation. Intuitively, it is insufficient not to distinguish the influence propagated between information source and target without directions. Hence, we formulate a bi-directional socialized factor graph model, which utilizes both the textual correlations between document pairs and the socialized augmentation networks behind the documents, such as user relationships and social interactions. These factors are modeled as attributes and dependencies among documents and their corresponding users, and then are distinguished on the direction level. We propose an effective learning algorithm to learn the proposed factor graph model with directions. Finally we propagate term counts to smooth documents based on the estimated influence. We run experiments on two instinctive datasets of Twitter and Weibo. The results validate the effectiveness of the proposed model. By incorporating direction information into the socialized language model smoothing, our approach obtains improvement over several alternative methods on both intrinsic and extrinsic evaluations measured in terms of perplexity, nDCG and MAP measurements.

MOTIVATION: Accurate detection of differentially expressed genes between tumor and normal samples is a primary approach of cancer-related biomarker identification. Due to the infiltration of tumor surrounding normal cells, the expression data derived from tumor samples would always be contaminated with normal cells. Ignoring such cellular contamination would deflate the power of detecting DE genes and further confound the biological interpretation of the analysis results. For the time being, there does not exists any differential expression analysis approach for RNA-seq data in literature that can properly account for the contamination of tumor samples. RESULTS: Without appealing to any extra information, we develop a new method 'contamDE' based on a novel statistical model that associates RNA-seq expression levels with cell types. It is demonstrated through simulation studies that contamDE could be much more powerful than the existing methods that ignore the contamination. In the application to two cancer studies, contamDE uniquely found several potential therapy and prognostic biomarkers of prostate cancer and non-small cell lung cancer. AVAILABILITY AND IMPLEMENTATION: An R package contamDE is freely available at http://homepage.fudan.edu.cn/zhangh/softwares/ CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Community question answering (CQA) has become an increasingly popular research topic. In this paper, we focus on the problem of question retrieval. Question retrieval in CQA can automatically find the most relevant and recent questions that have been solved by other users. However, the word ambiguity and word mismatch problems bring about new challenges for question retrieval in CQA. State-of-the-art approaches address these issues by implicitly expanding the queried questions with additional words or phrases using monolingual translation models. While useful, the effectiveness of these models is highly dependent on the availability of quality parallel monolingual corpora (e.g., question-answer pairs) in the absence of which they are troubled by noise issues. In this work, we propose an alternative way to address the word ambiguity and word mismatch problems by taking advantage of potentially rich semantic information drawn from other languages. Our proposed method employs statistical machine translation to improve question retrieval and enriches the question representation with the translated words from other languages via non-negative matrix factorization. Experiments conducted on real CQA data sets show that our proposed approach is promising.

Visualization is an important method in microbiome data analysis, and dimensionality reduction is a necessary procedure to achieve it. Multidimensional Scaling (MDS) is a popular method, which is necessary to compute the distance matrix. The Unifrac distance is very reasonable and biologically meaningful in the analysis of microbiome data. Due to the complexity of the phylogenetic tree and the high dimensionality of data, MDS needs a large amount of calculations to determine all the distances between pairs. In this paper, we proposed a novel dimensionality reduction algorithm based on Laplace matrix (DRLM) for the analysis of microbiome data. The experimental results indicate that both on synthesized and microbiome data, our algorithm DRLM can not only cluster the data more clearly, but also can significantly reduce the computational cost.

Microbial abundance dynamics along time axis can be used to explore complex interactions among microorganisms. This is very important to use time series data for understanding the structure and function of a microbial community and its dynamic characteristics with the purturbations of external environment and physiology. Species with Time Delay regulatory network of relationships will be more suitable for microbial interactions, because the regulation between microorganisms is often a slow process with delay, rather than an instantaneous process. In this study, a novel local linear manifold-constrained Vector Autoregression (LVAR) model that considered the time delay among microbial interactions is developed for analyzing microbiomics data in the application. The experimental results indicate that the new approach has better performance than several other VAR-based models and demonstrate its capability of extracting relevant microbial interactions.

Network is an exceptional way of depicting biological information. In biology, many different biological processes are represented by network, such as regulatory network, metabolic network and food web. In biology, network is a powerful supplement to the standard numerical data such as profile or count data. By absorbing network information, Vector autoregressive (VAR) model was proved to be an efficient approach to infer dynamic interactions in biological systems. Variants of network-regularized VAR with different penalties or regularization can avoid the problem of over-fitting and provide great potential in high-dimensional time series analysis. In this paper, we develop a novel regularization method for multivariate VAR which incorporates not only network topology but the signs of the network connections. By virtue of coordinate descent, we present a fast implementation for estimating model parameters. We then apply the proposed approach on several time series data sets especially a time series dataset of human gut microbiomes. The experimental results indicate that the new approach has better performance than other VAR-based models.

Video popularity prediction plays a foundational role in many aspects of life, such as recommendation systems and investment consulting. Because of its technological and economic importance, this problem has been extensively studied for years. However, four constraints have limited most related works' usability. First, most feature oriented models are inadequate in the social media environment, because many videos are published with no specific content features, such as a strong cast or a famous script. Second, many studies assume that there is a linear correlation existing between view counts from early and later days, but this is not the case in every scenario. Third, numerous works just take view counts into consideration, but discount associated sentiments. Nevertheless, it is the public opinions that directly drive a video's final success/failure. Also, many related approaches rely on a network topology, but such topologies are unavailable in many situations. Here, we propose a Dual Sentimental Hawkes Process (DSHP) to cope with all the problems above. DSHP's innovations are reflected in three ways: (1) it breaks the "Linear Correlation" assumption, and implements Hawkes Process; (2) it reveals deeper factors that affect a video's popularity; and (3) it is topology free. We evaluate DSHP on four types of videos: Movies, TV Episodes, Music Videos, and Online News, and compare its performance against 6 widely used models, including Translation Model, Multiple Linear Regression, KNN Regression, ARMA, Reinforced Poisson Process, and Univariate Hawkes Process. Our model outperforms all of the others, which indicates a promising application prospect.

Online social networks nowadays enjoy their worldwide prosperity, as they have revolutionized the way for people to discover, to share, and to distribute information. With millions of registered users and the proliferation of user-generated contents, the social networks become "giants", likely eligible to carry on any research tasks. However, the giants do have their Achilles Heel: extreme data sparsity. Compared with the massive data over the whole collection, individual posting documents, (e.g., a microblog less than 140 characters), seem to be too sparse to make a difference under various research scenarios, while actually they are different. In this paper we propose to tackle the Achilles Heel of social networks by smoothing the language model via influence propagation. We formulate a socialized factor graph model, which utilizes both the textual correlations between document pairs and the socialized augmentation networks behind the documents, such as user relationships and social interactions. These factors are modeled as attributes and dependencies among documents and their corresponding users. An efficient algorithm is designed to learn the proposed factor graph model. Finally we propagate term counts to smooth documents based on the estimated influence. Experimental results on Twitter and Weibo datasets validate the effectiveness of the proposed model. By leveraging the smoothed language model with social factors, our approach obtains significant improvement over several alternative methods on both intrinsic and extrinsic evaluations measured in terms of perplexity, nDCG and MAP results.

BACKGROUND: The identification of protein functional modules would be a great aid in furthering our knowledge of the principles of cellular organization. Most existing algorithms for identifying protein functional modules have a common defect -- once a protein node is assigned to a functional module, there is no chance to move the protein to the other functional modules during the follow-up processes, which lead the erroneous partitioning occurred at previous step to accumulate till to the end. RESULTS: In this paper, we design a new algorithm ADM (Adaptive Density Modularity) to detect protein functional modules based on adaptive density modularity. In ADM algorithm, according to the comparison between external closely associated degree and internal closely associated degree, the partitioning of a protein-protein interaction network into functional modules always evolves quickly to increase the density modularity of the network. The integration of density modularity into the new algorithm not only overcomes the drawback mentioned above, but also contributes to identifying protein functional modules more effectively. CONCLUSIONS: The experimental result reveals that the performance of ADM algorithm is superior to many state-of-the-art protein functional modules detection techniques in aspect of the accuracy of prediction. Moreover, the identified protein functional modules are statistically significant in terms of "Biological Process" annotated in Gene Ontology, which provides substantial support for revealing the principles of cellular organization.

BACKGROUND: Revealing protein complexes are important for understanding principles of cellular organization and function. High-throughput experimental techniques have produced a large amount of protein interactions, which makes it possible to predict protein complexes from protein-protein interaction (PPI) networks. However, the small amount of known physical interactions may limit protein complex detection. METHODS: The new PPI networks are constructed by integrating PPI datasets with the large and readily available PPI data from biomedical literature, and then the less reliable PPI between two proteins are filtered out based on semantic similarity and topological similarity of the two proteins. Finally, the supervised learning protein complex detection (SLPC), which can make full use of the information of available known complexes, is applied to detect protein complex on the new PPI networks. RESULTS: The experimental results of SLPC on two different categories yeast PPI networks demonstrate effectiveness of the approach: compared with the original PPI networks, the best average improvements of 4.76, 6.81 and 15.75 percentage units in the F-score, accuracy and maximum matching ratio (MMR) are achieved respectively; compared with the denoising PPI networks, the best average improvements of 3.91, 4.61 and 12.10 percentage units in the F-score, accuracy and MMR are achieved respectively; compared with ClusterONE, the start-of the-art complex detection method, on the denoising extended PPI networks, the average improvements of 26.02 and 22.40 percentage units in the F-score and MMR are achieved respectively. CONCLUSIONS: The experimental results show that the performances of SLPC have a large improvement through integration of new receivable PPI data from biomedical literature into original PPI networks and denoising PPI networks. In addition, our protein complexes detection method can achieve better performance than ClusterONE.

Microbial interactions play important roles on the structure and function of complex microbial communities. With the rapid accumulation of high-throughput metagenomic or 16S rRNA sequencing data, it is possible to infer complex microbial interactions. Co-occurrence patterns of microbial species among multiple samples are often utilized to infer interactions. There are few methods to consider the temporally interacting patterns among microbial species. In this paper, we present a Graph-regularized Vector Autoregressive (GVAR) model to infer causal relationships among microbial entities. The new model has advantage comparing to the original vector autoregressive (VAR) model. Specifically, GVAR can incorporate similarity information for microbial interaction inference--i.e., GVAR assumed that if two species are similar in the previous stage, they tend to have similar influence on the other species in the next stage. We apply the model on a time series dataset of human gut microbiome which was treated with repeated antibiotics. The experimental results indicate that the new approach has better performance than several other VAR-based models and demonstrate its capability of extracting relevant microbial interactions.

Disease-causing genes prioritization is very important for understanding mechanisms of diseases and biomedical applications, such as drug design. Previous studies have shown that promising candidate genes are mostly ranked according to their relatedness to known disease genes or closely related disease processes. Therefore, a dangling gene (isolated gene) with no edges in the network can not be effectively prioritized. These approaches tend to prioritize those genes that are highly connected in the PPI network and perform poorly when they are applied to loosely connected disease genes. Motivated by this observation, we propose a new disease-causing genes prioritization method that based on network diffusion and rank concordance (NDRC). The method is evaluated by leave-one-out cross validation on 1931 diseases in which at least one gene is known to be involved, and it is able to rank the true causal gene first in 849 of all 2542 cases, and as the experimental results suggest that NDRC significantly outperforms other existing methods such as RWR, VAVIEN, DADA and PRINCE on identifying loosely connected disease genes and which successfully put dangling genes as potential candidate disease genes. Furthermore, we apply NDRC method to study two representative diseases, Meckel syndrome 1 and Peroxisome biogenesis disorder 1A (Zellweger). Our study has also found that the complex disease-causing genes are divided into several modules that are closely associated with different disease phenotype.

The identification of modules in complex networks is important for the understanding of systems. Recent studies have shown those functional modules can be identified from the protein interaction a network, what's more, the complex modules have not only relatively high density, but also have high coefficient of affinity. However, these analyses are challenging because of the presence of unreliable interactions in PPT network. In this paper, in order to mine overlapping functional modules with various and effective biological characteristics, we propose a novel algorithm based on Connected Affinity and Multi-level Seed Extension (CAMSE). First, CAMSE integrates protein-protein interactions (PPI) with the protein-protein Connected Coefficient (CC) inferred from protein complexes collected in the MIPS database to enhance the modularization and biological character of the interaction network. Then we complete the seed selection, inner kernel extensions and outer extension to get core candidate function modules step by step. Finally, we integrated the modules with high repeat rate. The experimental results show that CAMSE can detect the functional modules much more effectively and accurately when it compared with other state-of-art algorithms CPM, CACE and IPC-MCE.

Microbiome datasets are often comprised of different representations or views which provide complementary information, such as metabolic pathways, taxonomic assignments and gene families. Computational methods for integration of multi-view information combine these data to create a comprehensive view of a given microbiome study. Similarity network fusion (SNF) provides a candidate to solve this problem by efficiently fusing similarity networks built from each data view into one network that represents the full spectrum of the underlying data. Based on this method, we propose a Robust Similarity Network Fusion (RSNF) approach which combines the strength of random forest to construct robust affine graph and the advantage of SNF at data aggregation. The experimental results indicate that the proposed strategy not only substantially outperforms single data type analysis but improve the clustering performance significantly comparing to several state-of-the-art methods in various datasets. The application on human microbiome data suggests that we can cluster microbiome samples in high accuracy.

Clinical notes are rich free-text data sources containing valuable symptom and medication information. Little research has been done on matching medication information with multiple symptoms information. Such a matching could provide valuable information for patients with multiple syndromes. We propose a Symptom-Medication (Symp-Med) matching framework to model symptom and medication relationships from clinical notes. After extracting symptom and medication concepts, we construct a weighted bipartite graph to represent the relationships between the two groups of concepts. The key is to efficiently answer user's symptom-medication queries using the graph. We formulate this problem as an Integer Linear Programming (ILP) problem. The objectives are to maximize the total edge weight and minimize the number of medication concepts. We first explore a Branch-and-Cut based algorithm. Then, we revise the combinational objective, and propose a Greedy-based algorithm for solving the Symp-Med problem. The Greedy-based algorithm performs better and significantly improves the computational costs.

We develop a novel distant supervised model that integrates the results from open information extraction techniques to perform relation extraction task from biomedical literature. Unlike state-of-the-art models for relation extraction in biomedical domain which are mainly based on supervised methods, our approach does not require manually-labeled instances. In addition, our model incorporates a grouping strategy to take into consideration the coordinating structure among entities co-occurred in one sentence. We apply our approach to extract gene expression relationship between genes and brain regions from literature. Results show that our methods can achieve promising performance over baselines of Transductive Support Vector Machine and with non-grouping strategy.

Dimensionality reduction and visualization are two important procedures in microbiome data analysis. With the intrinsic high dimensionality of the feature space in raw microbiome sequencing data, such as 16S rRNA, it requires proper simplification for possible further analysis. The explosively increasing size of data from large-scale microbiome studies inevitably and exponentially raises the computational complexity of existing algorithms, which is an urgent issue standing in the way requires addressing. This study proposed a new approach for dimensionality reduction and visualization on microbiome sequencing data associated with the very issue. This method not only greatly improves the efficiency of computing on microbiomic big data analysis by spectral interpolation technique but also preserves as much information as possible from original data with decent visualization results. With this adaptive method introduced to the large-scale studies of microbiome, we can better facilitate the revealing of patterns and insights of microbial communities.

In recent years, there are growing interests in developing novel approaches for inferring dynamic interactions in biological systems including gene transcription network and microbial interaction networks. Multivariate Vector Autoregression (MVAR) model is one of these efficient methods. Variants of MVAR with different penalties or regularizations can avoid the problem of over-fitting and provide great potential in high-dimensional data analysis. In this paper, we developed a novel regularization methods for MVAR via weighted fusion which consider the correlation among variables. The weighted fusion can potentially incorporate information redundancy among correlated variables for estimation and variable selection. Weighted fusion is also useful when the number of predictors p is larger than the number of observations n. In theory, we discuss the grouping effect of weighted fusion regularization for linear models. We then apply the proposed model on several time series data sets especially a time series dataset of human gut microbiomes. The experimental results indicate that the new approach has better performance that several other VAR-based models and we demonstrate its capability of extracting relevant microbial interactions.

Visualization of large-scale data is the first step to acquire preliminary insight into complex biological data. In recent years, many statistical visualization methods have been designed to support data visualization. Stochastic Neighbor Embedding (SNE) is one of these efficient approaches, which uses the probabilistic distance to model differences among data points within the data space. SNE and its variants (e.g. t-SNE) have demonstrated superiority over other methods in exploring complex data. By using these methods, however, similar data points tend to group together, which prevents the identification of subtle differences. A good visualization method should not only present clear data structure, but distinguish subtle differences. In this paper, we propose a novel extension of SNE. The approach has three innovations: (1) we replaced the Gaussian distribution in SNE with a Laplacian distribution on both high dimensional space and low dimensional space. The Laplace distribution has wider tails than the Gaussian distribution, and thus it can be used to overcome the over-crowding problem noted in SNE and its variants. (2) We used a symmetric modification of Kullback-Leibler divergence measure as the objective function which provides more flexibility to the model. (3) We add a graph Laplacian regularization terms to the objective function which have an advantage to preserve the manifold structure among data points. Experiments on simulation data and human microbiome data indicate that it has better visualization performance than other methods in distinguishing crowding data points.

The rapidly growing biomedical literature provides a significantly large and readily available source of PPI data. In this paper, we present supervised learning and data integration based complex detection approach. In this approach, a sophisticated natural language processing system, PPIExtractor, is employed to extract new PPI interactions from biomedical literature which are then integrated into original PPI networks. Then a supervised learning model, built by via of the information of available known complexes, is used in the multiple complex detection stages, e.g. the cliques filtering, growth, and candidate complex filtering. The experimental results on three yeast PPI networks demonstrate the effectiveness of our approach.

In this research, we develop a hybrid recommendation system for healthy living programs to patients with chronic diseases. Our experiments indicate that our model compared favorably against other real-world recommendation applications in terms of accuracy. We also demonstrated that the proposed hybrid algorithm performed better than traditional CF in terms of error rate, precision and recall.

To identify the underlying disease gene of human genetic disorders is a challenging and meaningful task in bioinformatics research. Recently, several methods we re developed based on PPI network, motivated by the observation that the disease genes of the same or similar diseases tend to lie close to each other in the PPI network. However, most of these methods based on the direct neighbors or shortest distance between disease genes, which ignore the global information of the PPI network. We develop a novel method for predicting disease gene based on function flow with PPI network. First, we map the known disease genes and candidate disease genes in linkage interval to PPI network. Then, simulate the process of function flow model which propagates information from the known disease genes to other genes (proteins) over the PPI network, and each protein gets a function score. Finally, candidate disease genes are ranked according to function score, the genes (proteins) with high score are considered as disease genes. The experimental results show our method is effective, and it can identify disease genes more accurately.

Topic modeling is a powerful tool to model documents to find their underlying topics. However, the unstructured nature of the raw text makes it hard to model the semantic relationship between the text units, which may be the words, phrases or sentences, and thus even harder to model their corresponding underlying topics. In our work, we try to examine the pairwise relationship of the underlying topics through relation extraction. We first extract the entity pairs within one relation tuple out of the raw text. Then, we model the relationship between the entity pairs by adding the dependencies between entities and their corresponding topics. We propose six different versions of Pairwise Topic Model (PTM) to simultaneously discover the latent topics and their pairwise relationship. The experiment on four data sets (AP news articles, DUC 2004 task2, Clinical Notes and Neuroscience Papers) shows the PTM models are better-structured language model than the traditional topic model Latent Dirichlet Allocation (LDA). Also, empirical results show that the proposed Pairwise Topic Models (PTMs) can explicitly explain how two topics are related.

Buzzwords are the main embodiment of Internet culture, which play an important role in public opinion analysis, social focus tracking and language evolution study. At present, questionnaire has been wildly used as a standard method to obtain network buzzwords, which is subjective and costly. In this paper, we will propose a novel algorithm relying on the time-distribution feature of words and a KL-divergence measure to estimate words' popularity so as to figure out buzzwords in a specific period. The time-distribution feature simply states the fact that buzzwords' usage has a sharp increase during a very short period, which is then modeled formally with the KL-divergence measure. Compared with traditional method involving much workforce, the automatic algorithm presented here is clearly more efficient. Moreover, buzzwords identified in this manner will not be affected by individual's subjective opinions, so they can reflect the language usage in practice better. When applying the algorithm to a social media big data set, our experimental results show that the proposed approach can accurately identify buzzwords in a certain period, which is highly coincident with results tagged manually.

BACKGROUND: In recent years, many protein complex mining algorithms, such as classical clique percolation (CPM) method and markov clustering (MCL) algorithm, have developed for protein-protein interaction network. However, most of the available algorithms primarily concentrate on mining dense protein subgraphs as protein complexes, failing to take into account the inherent organizational structure within protein complexes. Thus, there is a critical need to study the possibility of mining protein complexes using the topological information hidden in edges. Moreover, the recent massive experimental analyses reveal that protein complexes have their own intrinsic organization. METHODS: Inspired by the formation process of cliques of the complex social network and the centrality-lethality rule, we propose a new protein complex mining algorithm called Multistage Kernel Extension (MKE) algorithm, integrating the idea of critical proteins recognition in the Protein- Protein Interaction (PPI) network,. MKE first recognizes the nodes with high degree as the first level kernel of protein complex, and then adds the weighted best neighbour node of the first level kernel into the current kernel to form the second level kernel of the protein complex. This process is repeated, extending the current kernel to form protein complex. In the end, overlapped protein complexes are merged to form the final protein complex set. RESULTS: Here MKE has better accuracy compared with the classical clique percolation method and markov clustering algorithm. MKE also performs better than the classical clique percolation method both on Gene Ontology semantic similarity and co-localization enrichment and can effectively identify protein complexes with biological significance in the PPI network.

Accurately recognizing the object's behavior from the uncertain sensor data is a key issue of Internet of Things application. For example, in urban management monitoring system, it is necessary to have an autonomous analyzing module that can online monitor object's behavior based on environmental monitoring information in order to prevent an emergent situation in advance. In this work, we present an approximate object's behavior analysis method, called AOBA, which can recognize behavioral patterns of the hybrid objects which include patrolman, watering cart, street lamp etc. In intelligent urban management. AOBA consists of two phases: filtering phase and recognizing phase. In the filtering phase, a -approximate pre-matching algorithm based on q-grams distance is introduced to select possible pattern rapidly, which can discard huge amount insignificant or dirty data; in the recognizing phase, aiming to the temporal and the spatial characteristics of sensor data, an improved bit-parallel string matching algorithm is proposed to recognize the k-approximate multiple patterns over event sequences selected by the filtering phase. Experiments on real urban monitoring data and synthetic data show that the proposed method can efficiently discriminate object's behavior. Compared with the existing method, the proposed method provides a fault-tolerant approximate pattern recognition solution.

Microbial interaction, such as species competition and symbiotic relationships, plays important role to enable microorganisms to survive by establishing a homeostasis between microbial neighbors and local environments. Thanks to the recent accumulation of large-scale high-throughput sequencing data of complex microbial communities, there are increasing interests in identifying microbial interactions. Computational methods for microbial interactions inference are currently focused on the similarity among microbial individuals (i.e. cooccurrence and correlation patterns), however, less methods considered the dynamics of a single complex community over time. In this paper, we propose to use a multivariate statistical method - Multivariate Vector Autoregression (MVAR) to infer dynamic microbial interactions from the time series of human gut microbiomes. Specifically, we apply MVAR model on a time series data of human gut microbiomes which were treated with repeated antibiotics. The referred microbial interactions identify novel interactions which may provide a novel complementary to similarity or correlation-based methods.

Current data mining and statistical methods to extract patterns and relationships in microbiomic data are often based on several assumptions such as Euclidean, linear, continuous and metric space which may not be the true space of microbiomic data. For example, the microbial profiles (functional and taxonomic classifications) are often correlated in a hierarchical style. These assumptions prevent discovering the true relationships in microbiomic data analysis. Thus, it is urgent to develop new computational methods to overcome these assumptions and consider the microbiomic data properties in the analysis procedure. In this study, we will propose novel variable selection method based on manifold-constrained regularization (McRe). Considering the nonlinear and correlation structure of data, McRe get improved results in simulation data. The method is also applied to a microbiomic dataset.

A novel algorithm based on Connected Affinity Clique Extension (CACE) for mining overlapping functional modules in protein interaction network is proposed in this paper. In this approach, the value of protein connected affinity is interpreted as the reliability and possibility of interaction which is inferred from protein complexes. The protein interaction network is constructed as a weighted graph, and the weigh is dependent on the connected affinity coefficient. The experimental results of our CACE in two test data sets show that the CACE can detect the functional modules much more effective and accurate compared with other state-of-art algorithms CPM and IPC-MCE.

We develop an error detecting and tagging framework for reducing data entry errors in Electronic Medical Records (EMR) systems. We propose a taxonomy of data errors with three levels: Incorrect Format and Missing error, Out of Range error, and Inconsistent error. We aim to address the challenging problem of detecting erroneous input values that look statistically normal but are abnormal in medical sense. Detecting such an error needs to take patient medical history and population data into consideration. In particular, we propose a probabilistic method based on the assumption that the input value for a field depends on the historical records of this field, and is affected by other fields through dependency relationships. We evaluate our methods using the data collected from an EMR System. The results show that the method is promising for automatic data entry error detection.

The rising prevalence of Autism Spectrum Disorder (ASD) in the United States points to an increased need for services across the life span. Specialized services beginning at the earliest age possible are critical to maximizing long-term outcomes for children with ASD and their families. Many children later diagnosed with ASD will begin to receive services through the federally funded Early Intervention (EI) system that serves infants and toddlers from birth to age three. However, without formal recognition, services may not fully address the constellation of ASD symptoms. While ASD training in EI is becoming more widespread, there is still a need for better detection of ASD symptoms at younger ages. We hypothesized that initial EI assessment records which document the strengths and needs of children in EI, could be an important source for detecting ASD warning signs and aid state EI systems in earlier identification. In this research, we used EI records to evaluate classification techniques to identify suspected ASD cases. We improved the performance of machine learning techniques by developing and applying a unified ASD ontology to identify the most relevant features from EI records. The results indicate that using Support Vector Machine (SVM) with ontology-based unigrams as features yields the best performance. Our study shows that developing automatic approaches for quickly and effectively detecting suspected cases of ASD from non-standardized EI records earlier than most ASD cases are typically detected is promising.

Entity annotation is a fundamental step for entity relation extraction, information visualization and semantic web creation. Using domain specific ontologies to annotate a dataset for exploring, analyzing and integrating the entities and relations within the corpus is an important step for further study of the data. In this paper, we present an ontology based entity annotation system to annotate entities in neuroscience documents. To improve the annotation performance, we propose an ontology entity expansion method based on web service and ontology structure. We evaluate the proposed entity annotation method on real data obtained from Elsevier's BrainNavigator. The results show that using web service and ontology structure to expand ontology entities can improve the annotation result.

Density modularity can overcome this defect, but it use Simulated Annealing (SA) algorithm to search the maximal density modularity, which can't ensure to rapidly search the global optimal solution of problem. Based on this consideration, we propose a Closely Associated Degree (CAD) algorithm to discover protein functional module which continuously improve density modularity of PPI network. CAD first analyze the associated degree of protein node, then join it into the maximal associated degree module. When all modular structure are stable, CAD merges the pair of module that can bring the maximum increment of density modularity. This process is continually repeated that make density modularity to grow rapidly. Experimental results show that the CAD algorithm can effectively and accurately identify protein functional modules with biological significance in large-scale PPI network.

In order to overcome the limitations of global modularity and the deficiency of local modularity, we introduce a hybrid modularity measure LGQ (Local-Global Quantification) which adopts a suitable modularity adjustable parameter to control the balance of global detecting capability and local search capability in Protein-Protein Interaction (PPI) network. On the other hand, a new protein complex mining algorithm called BN-LGQ has been proposed, which integrates the definitions of best neighbor node and the modularity increment. And by comparison with other known algorithms, the experimental results show BN-LGQ performs a better accuracy on predicting protein complexes and has a higher match with the reference protein complexes. Moreover, it can identify protein complexes with better biological significance in PPI network.

Recent studies have shown that protein complex is composed of core and attachment proteins, and proteins inside the core are highly co-expressed. Based on this new concept, we reconstruct weighted PPI network by using gene expression data, and develop a novel protein complex identification algorithm from the angle of edge(PCIA-GeCo). First, we select the edge with high co-expressed coefficient as seed to form the preliminary cores. Then, the preliminary cores are filtered according to the weighted density of complex core to obtain the unique core. Finally, the protein complexes are generated by identifying attachment proteins for each core. A comprehensive comparison in term of F-measure, Coverage rate between our method and three other existing algorithms HUNTER, COACH and CORE has been made by comparing the predicted complexes against benchmark complexes. The evaluation results show our method PCIA-GeCo is effective; it can identify protein complexes more accurately.

We study the applications of hierarchical topic models to represent the content of website summaries. We concentrate on the DMOZ collection of Web extracts and propose a novel Tree Labeled LDA (tLLDA) algorithm to infer topic models using its manually compiled ontology. The algorithm takes advantage of the ontology structure and infers topic models by jointly modeling word and ontology node assignments for documents. We evaluate the performance of our topic modeling approach against that of four state-of-the-art algorithms (Labeled LDA, Hierarchically Labeled LDA, Hierarchically Supervised LDA and Supervised LDA) and show improvement in terms of perplexity and accuracy. Our evaluation shows that topic models produced by tLLDA outperform other algorithms in terms of perplexity for all test sets and all but one test case in terms of accuracy.

Sentiment analysis studies the public opinions towards an entity, and it is an important research area in data mining. Recently, a lot of sentiment analysis models have been proposed, including supervised and unsupervised approaches. However, the role of supervised models has been undermined by the phenomenon of big data, and the unsupervised ones are drawing more and more attention. But, most current unsupervised methods are based on Latent Dirichlet Allocation (LDA), and they need to specify the number of aspects in advance, making them subjective. In addition, these methods treat factual words and opinioned words the same, and assume that one sentence contains only one aspect, all of which make the existing unsupervised methods unsatisfactory. To solve these problems, this paper proposes a novel hybrid Hierarchical Dirichlet Process-Latent Dirichlet Allocation (HDP-LDA) model. This model can automatically determine the number of aspects, distinguish factual words from opinioned words, and further effectively extracts the aspect specific sentiment words. Experiment result shows that our model can clearly capture the aspects people mentioned and the specific sentiment words they use in each aspect, improving the performance of sentiment analysis efficiently. At last, we compared our model with the influential topic models, namely, JST, AUSM and Maxine-LDA, on the online restaurant review, and found our model performs very well.

In pervasive healthcare monitoring environments, data provenance, as one metadata, can help people analyze the reasons for medical accidents that are generated by complex events. This reasoning processing often encounters inaccurate time and irreversible reasoning problems. How to solve the uncertain process and fuzzy transformation time presents many challenges to the study of data provenance. In this paper, we propose a backward derivation model with the provenance semantic, backward fuzzy time reasoning net (BFTRN), to solve these two problems. We design a backward reasoning algorithm motivated by time automation theory based on this model. With regard to given life-critical alarms and some constraints, it cannot only derive all evolution paths and the possibility distribution of paths from historical information, but also efficiently compute the value of fuzzy time function for each transition of lift-critical complex alarms in the healthcare monitoring system. We also analyze the properties of BFTRN model in this paper. Experiments on real dataset show that the proposed model is efficient.

Forms are a major source of input for getting data into the underlying medical databases of electronic health/medical record (EHR/EMR) systems. Standardizing encounter forms and integrating data collected from different forms into a single database would greatly reduce heterogeneity. In this paper, we describe a framework, the fEHR-plus system, that annotates, maps, and integrates user-specified encounter forms into a single database. The development of the framework incorporates machine learning, standard medical terminology, and the principles of database design. We conduct an empirical study with 52 forms collected from 6 medical institutions for evaluating the performance of the fEHR-plus system. The overall results show that the system is promising towards improving interoperability among electronic health record systems.

Discovering the global structures of microbial community using large-scale metagenomes is a significant challenge in the era of post-genomics. Data-driven methods such as dimension reduction have shown to be useful when they applied on a metagenomics profile matrix which summarize the abundance of functional or taxonomic categorizations in metagenomic samples. Analogously, model-driven method such as probability topic model (PTM) has been used to build a generative model to simulate the generating of a microbial community based on metagenomic profiles. Data-driven methods are direct and simple, they provide intuitive visualization and understanding of metagenomic profiles. Model-driven methods are often complicated but give a generative mechanism of microbial community which is helpful in understanding the generating process of complex microbial ecology. However, results from model-driven methods are usually hard to visualize and there is less an intuitive understanding of them. We developed a new computational framework to incorporate the strength of data-driven methods into model-based methods and applied the framework to discover and interpret enterotype in human microbiome.

Using metagenomics to detect the global structure of microbial community remains a significant challenge. The structure of a microbial community and its functions are complicated not only because of the complex interactions among microbes but also their complicate interacting with confounding environmental factors. Recently dimension reduction methods such as Principle component analysis, Non-negative matrix factorization and Canonical correlation analysis have been employed extensively to investigate the complex structure embedded in metagenomic profiles which summarize the abundance of functional or taxonomic categorizations in metagenomic studies. However, metagenomic profiles are not necessary to meet the "Assumption of Linearity" behind these methods. Therefore it is worth to investigate how nonlinear methods can be utilized in metagenomic studies. In this paper, a nonlinear manifold learning method- Isomap is used to visualize and analyze large-scale metagenomic profiles. Isomap was applied on a large-scale Pfam profile which are derived from 45 metagenomes in Global Ocean Sampling expedition. In our result, a novel nonlinear structure of protein families is identified and the relationships among the identified nonlinear components and environmental factors of global ocean are explored. The results indicate the strength of nonlinear methods in learning the complex microbial structure. With the coming of the huge number of new sequenced metagenomes, nonlinear methods like Isomap could be necessary complementary tools to current widely used methods.

In this paper, we present a Bayesian-based Personalized Laboratory Tests prediction (BPLT) model to solve a real world medical problem: how to recommend laboratory tests to a group of patients? Given a patient who has conducted several laboratory tests, BPLT model recommends further laboratory tests that are the most related to this patient. We regard this laboratory test prediction problem as a special classification problem, where a new laboratory test belongs to either a "taken" or "not-taken" class. Our goal is to find the laboratory tests with high probability of "taken" and low probability of "not taken". Based on Bayesian method, the BPLT model builds a weighting function to investigate the correlations among laboratory tests and generate the rank of laboratory tests. In order to evaluate the proposed BPLT model, we further propose a novel evaluation metric to subjectively measure the accuracy of BPLT model. Experimental results show that BPLT model achieves good performance on the real data sets and provides a good solution to our real world application.

In order to realize the Semantic Web, various structures on the Web including Web forms need to be annotated with and mapped to domain ontologies. We present a machine learning-based automatic approach for discovering complex mappings from Web forms to ontologies. A complex mapping associates a set of semantically related elements on a form to a set of semantically related elements in an ontology. Existing schema mapping solutions mainly rely on integrity constraints to infer complex schema mappings. However, it is difficult to extract rich integrity constraints from forms. We show how machine learning techniques can be used to automatically discover complex mappings between Web forms and ontologies. The challenge is how to capture and learn the complicated knowledge encoded in existing complex mappings. We develop an initial solution that takes a naive Bayesian approach. We evaluated the performance of the solution on various domains. Our experimental results show that the solution returns the expected mappings as the top-1 results usually among several hundreds candidate mappings for more than 80% of the test cases. Furthermore, the expected mappings are always returned as the top-k results with k<4. The experiments have demonstrated that the approach is effective and has the potential to save significant human efforts.

In this paper, we deal with two research issues: the automation of visual attribute identification and semantic relation learning between visual attributes and object categories. The contribution is two-fold, firstly, we provide uniform framework to reliably extract both categorical attributes and depictive attributes. Secondly, we incorporate the obtained semantic associations between visual attributes and object categories into a text-based topic model and extract descriptive latent topics from external textual knowledge sources. Specifically, we show that in mining natural language descriptions from external knowledge sources, the relation between semantic visual attributes and object categories can be encoded as Must-Links and Cannot-Links, which can be represented by Dirichlet-Forest prior. To alleviate the workload of manual supervision and labeling in image categorization process, we introduce a semi-supervised training framework using soft-margin semi-supervised SVM classifier. We also show that the large-scale image categorization results can be significantly improved by combining automatically acquired visual attributes. Experimental results show that the proposed model achieves better ability in describing object-related attributes and makes the inferred latent topics more descriptive.

This paper presents a tag-topic model with Dirichlet Forest prior (TTM-DF) for semantic knowledge acquisition from blog. The TTM-DF model extends the tag-topic model (TTM) by replacing the Dirichlet prior with the Dirichlet Forest prior over the topic-word multinomial. The correlation between words are calculated to generate a set of Must-Links and Cannot-Links, then the structures of Dirichlet trees are obtained though encoding the constraints of Must-Links and Cannot-Links. Words under the same subtrees are expected to be more correlated than words under different subtrees. We conduct experiments on a synthetic and a blog dataset. Both of the experimental results show that the TTM-DF model performs much better than the TTM model. It can improve the coherence of the underlying topics and the tag-topic distributions, and capture semantic knowledge effectively.

This paper proposes a novel topic model, Author-Conference Topic-Connection (ACTC) Model for academic network search. The ACTC Model extends the author-conference-topic (ACT) model by adding subject of the conference and the latent mapping information between subjects and topics. It simultaneously models topical aspects of papers, authors and conferences with two latent topic layers: a subject layer corresponding to conference topic, and a topic layer corresponding to the word topic. Each author would be associated with a multinomial distribution over subjects of conference (eg., KM, DB, IR for CIKM 2012), the conference(CIKM 2012), and the topics are respectively generated from a sampled subject. Then the words are generated from the sampled topics. We conduct experiments on a data set with 8,523 authors, 22,487 papers and 1,243 conferences from the well-known Arnetminer website, and train the model with different number of subjects and topics. For a qualitative evaluation, we compare ACTC with three others models LDA, Author-Topic (AT) and ACT in academic search services. Experiments show that ACTC can effectively capture the semantic connection between different types of information in academic network and perform well in expert searching and conference searching.

BACKGROUND: The growth of the biomedical information requires most information retrieval systems to provide short and specific answers in response to complex user queries. Semantic information in the form of free text that is structured in a way makes it straightforward for humans to read but more difficult for computers to interpret automatically and search efficiently. One of the reasons is that most traditional information retrieval models assume terms are conditionally independent given a document/passage. Therefore, we are motivated to consider term associations within different contexts to help the models understand semantic information and use it for improving biomedical information retrieval performance. RESULTS: We propose a term association approach to discover term associations among the keywords from a query. The experiments are conducted on the TREC 2004-2007 Genomics data sets and the TREC 2004 HARD data set. The proposed approach is promising and achieves superiority over the baselines and the GSP results. The parameter settings and different indices are investigated that the sentence-based index produces the best results in terms of the document-level, the word-based index for the best results in terms of the passage-level and the paragraph-based index for the best results in terms of the passage2-level. Furthermore, the best term association results always come from the best baseline. The tuning number k in the proposed recursive re-ranking algorithm is discussed and locally optimized to be 10. CONCLUSIONS: First, modelling term association for improving biomedical information retrieval using factor analysis, is one of the major contributions in our work. Second, the experiments confirm that term association considering co-occurrence and dependency among the keywords can produce better results than the baselines treating the keywords independently. Third, the baselines are re-ranked according to the importance and reliance of latent factors behind term associations. These latent factors are decided by the proposed model and their term appearances in the first round retrieved passages.

In this paper, we present a method that enable both homology-based approach and composition-based approach to further study the functional core (i.e., microbial core and gene core, correspondingly). In the proposed method, the identification of major functionality groups is achieved by generative topic modeling, which is able to extract useful information from unlabeled data. We first show that generative topic model can be used to model the taxon abundance information obtained by homology-based approach and study the microbial core. The model considers each sample as a “document,” which has a mixture of functional groups, while each functional group (also known as a “latent topic”) is a weight mixture of species. Therefore, estimating the generative topic model for taxon abundance data will uncover the distribution over latent functions (latent topic) in each sample. Second, we show that, generative topic model can also be used to study the genome-level composition of “N-mer” features (DNA subreads obtained by composition-based approaches). The model consider each genome as a mixture of latten genetic patterns (latent topics), while each functional pattern is a weighted mixture of the “N-mer” features, thus the existence of core genomes can be indicated by a set of common N-mer features. After studying the mutual information between latent topics and gene regions, we provide an explanation of the functional roles of uncovered latten genetic patterns. The experimental results demonstrate the effectiveness of proposed method.

With the fast growing number of works utilizing link information in enhancing unsupervised document clustering, it is becoming necessary to make a comparative evaluation of the impacts of different link types on document clustering. Various types of links between text documents, including explicit links such as citation links and hyperlinks, implicit links such as coauthorship and cocitation links, and similarity links such as content similarity links, convey topic similarity or topic transferring patterns, which is very useful for document clustering. In this paper, we adopt a clustering algorithm based on Markov random field and relaxation labeling, which employs both content and linkage information, to evaluate the effectiveness of the aforementioned types of links for document clustering on ten data sets. The experimental results show that linkage information is quite effective in improving content-based document clustering. Furthermore, a series of important findings regarding the impacts of different link types on document clustering is discovered through our experiments.

In this paper, based on the functional elements derived from non-redundant CDs catalogue, we show that the configuration of functional groups in meta-genome samples can be inferred by probabilistic topic modeling. The probabilistic topic modeling is a Bayesian method that is able to extract useful topical information from unlabeled data. When used to study microbial samples (assuming that relative abundance of functional elements is already obtained by a homology-based approach), each sample can be considered as a 'document', which has a mixture of functional groups, while each functional group (also known as a 'latent topic') is a weight mixture of functional elements (including taxonomic levels, and indicators of gene orthologous groups and KEGG pathway mappings). The functional elements bear an analogy with 'words'. Estimating the probabilistic topic model can uncover the configuration of functional groups (the latent topic) in each sample. The experimental results demonstrate the effectiveness of our proposed method.

In this paper, we propose an approach based on a topic generative model called Latent Dirichlet Allocation (LDA) to promoting ranking diversity for biomedical information retrieval. Different from other approaches or models which consider aspects on word level, our approach assumes that aspects should be identified by the topics of retrieved documents. We present LDA model to discover topic distribution of retrieval passages and word distribution of each topic dimension, and then re-rank retrieval results with topic distribution similarity between passages based on JV-size slide window. Experiments on TREC 2007 Genomics collection and two distinctive IR baseline runs demonstrate the effectiveness of our method in promoting ranking diversity for biomedical information retrieval. Evaluation results show that our approach can achieve 8% improvement over the highest Aspect MAP reported in TREC 2007 Genomics track.

In silico approaches for drug-induced toxicity evaluation are likely to enable improved screening of new chemical entities during drug discovery and development. Analysis of protein-protein interaction (PPI) networks using an edge centrality-based measure has previously been shown to reveal protein modules that may be associated with drug-induced toxicity. Here, we extend the algorithm by integrating protein interaction information with in vitro gene expression data from tissue treated with drugs known to be associated with the toxicity. We evaluate the new measure for its ability to detect non- immune neutropenia related proteins and propose a biomarker panel that may be valuable for screening future drug candidates.

Modeling and mining term association is important for information retrieval, which allows an information retrieval system to retrieve relevant documents/passages more precisely. In this paper, we propose a new approach for discovering term associations among the keywords from a query. First, factor analysis is applied to discover some hidden common factors as the "eliteness" variables that can be used to estimate the importance of term associations. Second, a factor analysis based model and a corresponding algorithm are proposed. Then, we report experimental results that confirm the effectiveness and superiority of the proposed term association approach. Our approach achieves excellent results on the TREC 2007 and 2006 data sets, which provides a promising avenue for constructing high performance information retrieval systems.

In this paper, we proposed a perspective Hierarchical Dirichlet Process (pHDP) model to deal with user-tagged image modeling. The contribution is two-fold. Firstly, we associate image features with image tags. Secondly, we incorporate the user's perspectives into the image tag generation process and introduce new latent variables to determine if an image tag is generated from user's perspectives or from the image content. Therefore, the model is able to extract both embedded semantic components and user's perspectives from user-tagged images. Based on the proposed pHDP model, we achieve automatic image tagging with users' perspective. Experimental results show that the pHDP model achieves better image tagging performance compared to state-of-the-art topic models.

BACKGROUND: The wide use of high-throughput DNA microarray technology provide an increasingly detailed view of human transcriptome from hundreds to thousands of genes. Although biomedical researchers typically design microarray experiments to explore specific biological contexts, the relationships between genes are hard to identified because they are complex and noisy high-dimensional data and are often hindered by low statistical power. The main challenge now is to extract valuable biological information from the colossal amount of data to gain insight into biological processes and the mechanisms of human disease. To overcome the challenge requires mathematical and computational methods that are versatile enough to capture the underlying biological features and simple enough to be applied efficiently to large datasets. METHODS: Unsupervised machine learning approaches provide new and efficient analysis of gene expression profiles. In our study, two unsupervised knowledge-based matrix factorization methods, independent component analysis (ICA) and nonnegative matrix factorization (NMF) are integrated to identify significant genes and related pathways in microarray gene expression dataset of Alzheimer's disease. The advantage of these two approaches is they can be performed as a biclustering method by which genes and conditions can be clustered simultaneously. Furthermore, they can group genes into different categories for identifying related diagnostic pathways and regulatory networks. The difference between these two method lies in ICA assume statistical independence of the expression modes, while NMF need positivity constrains to generate localized gene expression profiles. RESULTS: In our work, we performed FastICA and non-smooth NMF methods on DNA microarray gene expression data of Alzheimer's disease respectively. The simulation results shows that both of the methods can clearly classify severe AD samples from control samples, and the biological analysis of the identified significant genes and their related pathways demonstrated that these genes play a prominent role in AD and relate the activation patterns to AD phenotypes. It is validated that the combination of these two methods is efficient. CONCLUSIONS: Unsupervised matrix factorization methods provide efficient tools to analyze high-throughput microarray dataset. According to the facts that different unsupervised approaches explore correlations in the high-dimensional data space and identify relevant subspace base on different hypotheses, integrating these methods to explore the underlying biological information from microarray dataset is an efficient approach. By combining the significant genes identified by both ICA and NMF, the biological analysis shows great efficient for elucidating the molecular taxonomy of Alzheimer's disease and enable better experimental design to further identify potential pathways and therapeutic targets of AD.

BACKGROUND: In the biomedical domain, the desired information of a question (query) asked by biologists usually is a list of a certain type of entities covering different aspects that are related to the question, such as genes, proteins, diseases, mutations, etc. Hence it is important for a biomedical information retrieval system to be able to provide comprehensive and diverse answers to fulfill biologists' information needs. However, traditional retrieval models assume that the relevance of a document is independent of the relevance of other documents. This assumption may result in high redundancy and low diversity in the retrieval ranked lists. RESULTS: In this paper, we propose a relevance-novelty combined model, named RelNov model, based on the framework of an undirected graphical model. It consists of two component models, namely the aspect-term relevance model and the aspect-term novelty model. They model the relevance of a document and the novelty of a document respectively. We show that our approach can achieve 16.4% improvement over the highest aspect level MAP reported in the TREC 2007 Genomics track, and 9.8% improvement over the highest passage level MAP reported in the TREC 2007 Genomics track. CONCLUSIONS: The proposed combination model which models aspects, terms, topic relevance and document novelty as potential functions is demonstrated to be effective in promoting ranking diversity as well as in improving relevance of ranked lists for genomics search. We also show that the use of aspect plays an important role in the model. Moreover, the proposed model can integrate various different relevance and novelty measures easily.

Feature representation is essential to machine learning and text mining. In this paper, we present a feature coupling generalization (FCG) framework for generating new features from unlabeled data. It selects two special types of features, i.e., example-distinguishing features (EDFs) and class-distinguishing features (CDFs) from original feature set, and then generalizes EDFs into higher-level features based on their coupling degrees with CDFs in unlabeled data. The advantage is: EDFs with extreme sparsity in labeled data can be enriched by their co-occurrences with CDFs in unlabeled data so that the performance of these low-frequency features can be greatly boosted and new information from unlabeled can be incorporated. We apply this approach to three tasks in biomedical literature mining: gene named entity recognition (NER), protein-protein interaction extraction (PPIE), and text classification (TC) for gene ontology (GO) annotation. New features are generated from over 20 GB unlabeled PubMed abstracts. The experimental results on BioCreative 2, AIMED corpus, and TREC 2005 Genomics Track show that 1) FCG can utilize well the sparse features ignored by supervised learning. 2) It improves the performance of supervised baselines by 7.8 percent, 5.0 percent, and 5.8 percent, respectively, in the tree tasks. 3) Our methods achieve 89.1, 64.5 F-score, and 60.1 normalized utility on the three benchmark data sets.

Social tagging is a major characteristic of Web 2.0. A social tagging system can be modeled with a tripartite network of users, resources, and tags. In this paper, we investigate how to enhance Web clustering by leveraging the tripartite network of social tagging systems. We propose a clustering method called “Tripartite Clustering” which clusters the three types of nodes (resources, users, and tags) simultaneously by only utilizing the links in the social tagging network. We also investigate two other approaches to exploit social tagging for clustering with K-means and Link K-means. All the clustering methods are experimented on a real-world social tagging data set sampled from del.icio.us. The clustering results are evaluated against a human-maintained Web directory. The experimental results show that the social tagging network is a very useful information source for document clustering. All social-annotation-based clustering methods can significantly improve the performance of content-based clustering. Compared to social-annotation-based K-means and Link K-means, Tripartite Clustering achieves equivalent or better performance and produces more useful information.

Recently, the concept of a species containing both core and distributed genes, known as the supra- or pangenome theory, has been introduced. In this paper, we aim to develop a new method that is able to analyze the genome-level composition of DNA sequences, in order to characterize a set of common genomic features shared by the same species and tell their functional roles. To achieve this end, we firstly apply a composition-based approach to break down DNA sequences into sub-reads called the `N-mer' and represent the sequences by N-mer frequencies. Then, we introduce the Latent Dirichlet Allocation (LDA) model to study the genome-level statistic patterns (a.k.a. latent topics) of the `N-mer' features. Each estimated latent topic represents a certain component of the whole genome. With the help of the BioJava toolkit, we access to the gene region information of reference sequences from the NCBI database. We use our data mining framework to investigate two areas: 1) do strains within species share similar core and distributed topics? and 2) do genes with similar functional roles contain similar latent topics? After studying the mutual information between latent topics and gene regions, we provide examples of each, where the BioCyc database is used to correlate pathway and reaction information to the genes. The examples demonstrate the effectiveness of proposed method.

Unsupervised machine learning approaches are efficient analysis tools for DNA microarray technique which can accumulate hundreds of thousands of genes expression levels in a single experiment. In our study, two unsupervised knowledge-based matrix factorization methods, independent component analysis (ICA) and nonnegative matrix factorization (NMF) are explored to identify significant genes and related pathways in microarray gene expression dataset. The advantage of these two approaches is they can be performed as a biclustering method by which genes and conditions can be clustered simultaneously. Furthermore, they can group genes into different categories for identifying related diagnostic pathways and regulatory networks. The difference between these two method lies in ICA assume statistical independence of the expression modes, while NMF need positivity constrains to generate localized gene expression profiles. By combining the significant genes identified by both ICA and NMF, the simulation results show great efficient for finding underlying biological processes and related pathways in Alzheimer's disease (AD) and the activation patterns to AD phenotypes.

Traditional retrieval models assume that the relevance of a document is independent of the relevance of other documents. However, this assumption may result in high redundancy and low diversity in a ranked list. In order to provide comprehensive and diverse answers to fulfill biologists' information need, we propose a relevance-novelty combined model, named RelNov model, based on the framework of an undirected graphical model. Experiments conducted on the TREC 2006 and 2007 Genomics collections show that the proposed approach is effective in promoting both diversity and relevance of retrieval ranked lists.

The explosive increase of image data on Internet has made it an important, yet very challenging task to index and automatically annotate image data. To achieve that end, sophisticated algorithms and models have been proposed to study the correlation between image content and corresponding text description. Despite the success of previous works, however, researchers are still facing two major difficulties that may undermine their effort of providing reliable and accurate annotations for images. The first difficulty is lacking of comprehensive benchmark image dataset with high quality text descriptions. The second difficulty is lacking of effective way to represent the image content and make it associate with the text descriptions. In our paper, we aim to deal with both problems. To deal with the first problem, we utilize Wikipedia as external knowledge source and enrich the ontology structure of ImageNet database with comprehensive and highly-reliable text descriptions from Wikipedia articles. To address the second problem, we develop a Probabilistic Topic-Connection (PTC) model to represent the connection between latent semantic topic in text description and latent patterns from image feature space. We compare the performance of our model with the currently popular Correspondence LDA (Corr-LDA) model under the same automatic image annotation scenario using cross-validation. Experimental results demonstrate that our model is able to well represent the connection between latent semantic topics and latent patterns in image feature space, thus facilitates knowledge organization and understanding of both image and text descriptions.

In this paper, we propose a new probabilistic generative model, called Topic-Perspective Model, for simulating the generation process of social annotations. Different from other generative models, in our model, the tag generation process is separated from the content term generation process. While content terms are only generated from resource topics, social tags are generated by resource topics and user perspectives together. The proposed probabilistic model can produce more useful information than any other models proposed before. The parameters learned from this model include: (1) the topical distribution of each document, (2) the perspective distribution of each user, (3) the word distribution of each topic, (4) the tag distribution of each topic, (5) the tag distribution of each user perspective, (6) and the probabilistic of each tag being generated from resource topics or user perspectives. Experimental results show that the proposed model has better generalization performance or tag prediction ability than other two models proposed in previous research.

We present NegativeRank, a novel graph-based sentence ranking model to improve the diversity of focused summary by performing random walks over sentence graph with negative edge weights. Unlike the typical eigenvector centrality ranking, our method models the redundancy among sentence nodes as the negative edges. The negative edges can be thought of as the propagation of disapproval votes which can be used to penalize redundant sentences. As the iterative process continues, the initial ranking score of a given node will be adjusted according to a long-term negative endorsement from other sentence nodes. The evaluation results confirm that our proposed method is very effective in improving the diversity of the focused summary, compared to several well-known text summarization methods.

BACKGROUND: Extracting protein-protein interactions from biomedical literature is an important task in biomedical text mining. Supervised machine learning methods have been used with great success in this task but they tend to suffer from data sparseness because of their restriction to obtain knowledge from limited amount of labelled data. In this work, we study the use of unlabeled biomedical texts to enhance the performance of supervised learning for this task. We use feature coupling generalization (FCG) - a recently proposed semi-supervised learning strategy - to learn an enriched representation of local contexts in sentences from 47 million unlabeled examples and investigate the performance of the new features on AIMED corpus. RESULTS: The new features generated by FCG achieve a 60.1 F-score and produce significant improvement over supervised baselines. The experimental analysis shows that FCG can utilize well the sparse features which have little effect in supervised learning. The new features perform better in non-linear classifiers than linear ones. We combine the new features with local lexical features, obtaining an F-score of 63.5 on AIMED corpus, which is comparable with the current state-of-the-art results. We also find that simple Boolean lexical features derived only from local contexts are able to achieve competitive results against most syntactic feature/kernel based methods. CONCLUSIONS: FCG creates a lot of opportunities for designing new features, since a lot of sparse features ignored by supervised learning can be utilized well. Interestingly, our results also demonstrate that the state-of-the art performance can be achieved without using any syntactic information in this task.

BACKGROUND: Protein translation is a vital cellular process for any living organism. The availability of interaction databases provides an opportunity for researchers to exploit the immense amount of data in silico such as studying biological networks. There has been an extensive effort using computational methods in deciphering the transcriptional regulatory networks. However, research on translation regulatory networks has caught little attention in the bioinformatics and computational biology community. RESULTS: In this paper, we present an exploratory analysis of yeast protein translation regulatory networks using hierarchical random graphs. We derive a protein translation regulatory network from a protein-protein interaction dataset. Using a hierarchical random graph model, we show that the network exhibits well organized hierarchical structure. In addition, we apply this technique to predict missing links in the network. CONCLUSIONS: The hierarchical random graph mode can be a potentially useful technique for inferring hierarchical structure from network data and predicting missing links in partly known networks. The results from the reconstructed protein translation regulatory networks have potential implications for better understanding mechanisms of translational control from a system's perspective.

The novel connection between Raynaud disease and fish oils was uncovered from two disjointed biomedical literature sets by Swanson in 1986. Since then, there have been many approaches to uncover novel connections by mining the biomedical literature. One of the popular approaches is to adapt the association rule (AR) method to automatically identify implicit novel connections between concept A and concept C from two disjointed sets of documents through intermediate B concept. Since A and C concepts do not occur together in the same data set, the mining goal is to find novel connection among A and C concepts in the disjoint data sets. It first applies association rule to the two disjointed biomedical literature sets separately to generate two rule sets (A→B, B→C), and then applies transitive law to get the novel connections A→C. However, this approach generates a huge number of possible connections among the millions of biomedical concepts and a lot of these hypothetical connections are spurious, useless, and/or biologically meaningless. Thus it is essential to develop new approach to generate highly likely novel and biologically relevant connections among the biomedical concepts. This paper presents a biomedical semantic-based association rule system (Bio-SARS) that significantly reduce spurious/useless/biologically irrelevant connections through semantic filtering. Compared to other approaches such as latent semantic indexing and traditional association rule-based approach, our approach generates much fewer rules and a lot of these rules represent relevant connections among biological concepts. © 2009 Wiley Periodicals, Inc.

This paper describes a round-trip engineering approach for incrementally maintaining mappings between conceptual models and relational schemas. When either schema or conceptual model evolves to accommodate new information needs, the existing mapping must be maintained accordingly to continuously provide valid services. In this paper, the authors examine the mappings specifying “consistent” relationships between models. First, they define the consistency of a conceptual-relational mapping through “semantically compatible” instances. Next, the authors analyze the knowledge encoded in the standard database design process and develop round-trip algorithms for incrementally maintaining the consistency of conceptual-relational mappings under evolution. Finally, they conduct a set of comprehensive experiments. The results show that the proposed solution is efficient and provides significant benefits in comparison to the mapping reconstructing approach.

Biclustering methods allow us to identify genes with similar behavior with respect to different conditions. Ant colony optimization (ACO) algorithms have been shown to be effective problem solving strategies for multiple objective optimization (MOO). Multiple objective ant colony optimization (MOACO) mainly focuses on solving the multiple objective combinatorial optimization problems. This paper incorporates crowding update technology into MOACOB and proposes a novel crowding based MOACO biclustering algorithm to mine biclusters from microarray dataset. Experimental results are shown for biclustering algorithm on two real gene expression dataset.

Protein translation is a vital cellular process for any living organism. The availability of interaction databases provides an opportunity for researchers to exploit the immense amount of data in silico such as studying biological networks. There has been an extensive effort using computational methods in deciphering the transcriptional regulatory networks. However, research on translation regulatory networks has caught little attention in the bioinformatics and computational biology community. In this paper, we present an exploratory analysis of yeast protein translation regulatory networks using hierarchical random graphs. We derive a protein translation regulatory network from a protein-protein interaction dataset. Using a hierarchical random graph model, we show that the network exhibits well organized hierarchical structure. In addition, we apply this technique to predict missing links in the network. The results have potential implications for better understanding mechanisms of translational control from a system's perspective.

In traditional text clustering methods, documents are represented as "bags of words" without considering the semantic information of each document. For instance, if two documents use different collections of core words to represent the same topic, they may be falsely assigned to different clusters due to the lack of shared core words, although the core words they use are probably synonyms or semantically associated in other forms. The most common way to solve this problem is to enrich document representation with the background knowledge in an ontology. There are two major issues for this approach: (1) the coverage of the ontology is limited, even for WordNet or Mesh, (2) using ontology terms as replacement or additional features may cause information loss, or introduce noise. In this paper, we present a novel text clustering method to address these two issues by enriching document representation with Wikipedia concept and category information. We develop two approaches, exact match and relatedness-match, to map text documents to Wikipedia concepts, and further to Wikipedia categories. Then the text documents are clustered based on a similarity metric which combines document content information, concept information as well as category information. The experimental results using the proposed clustering framework on three datasets (20-newsgroup, TDT2, and LA Times) show that clustering performance improves significantly by enriching document representation with Wikipedia concepts and categories.

BACKGROUND: High-throughput microarray technologies have generated and accumulated massive amounts of gene expression datasets that contain expression levels of thousands of genes under hundreds of different experimental conditions. The microarray datasets are usually presented in 2D matrices, where rows represent genes and columns represent experimental conditions. The analysis of such datasets can discover local structures composed by sets of genes that show coherent expression patterns under subsets of experimental conditions. It leads to the development of sophisticated algorithms capable of extracting novel and useful knowledge from a biomedical point of view. In the medical domain, these patterns are useful for understanding various diseases, and aid in more accurate diagnosis, prognosis, treatment planning, as well as drug discovery. RESULTS: In this work we present the CMOPSOB (Crowding distance based Multi-objective Particle Swarm Optimization Biclustering), a novel clustering approach for microarray datasets to cluster genes and conditions highly related in sub-portions of the microarray data. The objective of biclustering is to find sub-matrices, i.e. maximal subgroups of genes and subgroups of conditions where the genes exhibit highly correlated activities over a subset of conditions. Since these objectives are mutually conflicting, they become suitable candidates for multi-objective modelling. Our approach CMOPSOB is based on a heuristic search technique, multi-objective particle swarm optimization, which simulates the movements of a flock of birds which aim to find food. In the meantime, the nearest neighbour search strategies based on crowding distance and -dominance can rapidly converge to the Pareto front and guarantee diversity of solutions. We compare the potential of this methodology with other biclustering algorithms by analyzing two common and public datasets of gene expression profiles. In all cases our method can find localized structures related to sets of genes that show consistent expression patterns across subsets of experimental conditions. The mined patterns present a significant biological relevance in terms of related biological processes, components and molecular functions in a species-independent manner. CONCLUSION: The proposed CMOPSOB algorithm is successfully applied to biclustering of microarray dataset. It achieves a good diversity in the obtained Pareto front, and rapid convergence. Therefore, it is a useful tool to analyze large microarray datasets.

It is well known that Affymetrix microarrays are widely used to predict genome-wide gene expression and genome-wide genetic polymorphisms from RNA and genomic DNA hybridization experiments, respectively. It has recently been proposed to integrate the two predictions by use of RNA microarray data only. Although the ability to detect single feature polymorphisms (SFPs) from RNA microarray data has many practical implications for genome study in both sequenced and unsequenced species, it raises enormous challenges for statistical modelling and analysis of microarray gene expression data for this objective. Several methods are proposed to predict SFPs from the gene expression profile. However, their performance is highly vulnerable to differential expression of genes. The SFPs thus predicted are eventually a reflection of differentially expressed genes rather than genuine sequence polymorphisms. To address the problem, we developed a novel statistical method to separate the binding affinity between a transcript and its targeting probe and the parameter measuring transcript abundance from perfect-match hybridization values of Affymetrix gene expression data. We implemented a Bayesian approach to detect SFPs and to genotype a segregating population at the detected SFPs. Based on analysis of three Affymetrix microarray datasets, we demonstrated that the present method confers a significantly improved robustness and accuracy in detecting the SFPs that carry genuine sequence polymorphisms when compared to its rivals in the literature. The method developed in this paper will provide experimental genomicists with advanced analytical tools for appropriate and efficient analysis of their microarray experiments and biostatisticians with insightful interpretation of Affymetrix microarray data.

In this paper, we present a novel method to mine, model, and evaluate a regulatory system executing cellular functions that can be represented as a biomolecular network. Our method consists of two steps. First, a novel scale-free network clustering approach is applied to such a biomolecular network to obtain various subnetworks. Second, computational models are generated for the subnetworks and simulated to predict their behavior in the cellular context. We discuss and evaluate some of the advanced computational modeling approaches, in particular, state-space modeling, probabilistic Boolean network modeling, and fuzzy logic modeling. The modeling and simulation results represent hypotheses that are tested against high-throughput biological datasets (microarrays and/or genetic screens) under normal and perturbation conditions. Experimental results on time-series gene expression data for the human cell cycle indicate that our approach is promising for subnetwork mining and simulation from large biomolecular networks.

Generating high-quality gene clusters and identifying the underlying biological mechanism of the gene clusters are the important goals of clustering gene expression analysis. To get high-quality cluster results, most of the current approaches rely on choosing the best cluster algorithm, in which the design biases and assumptions meet the underlying distribution of the dataset. There are two issues for this approach: 1) usually, the underlying data distribution of the gene expression datasets is unknown and 2) there are so many clustering algorithms available and it is very challenging to choose the proper one. To provide a textual summary of the gene clusters, the most explored approach is the extractive approach that essentially builds upon techniques borrowed from the information retrieval, in which the objective is to provide terms to be used for query expansion, and not to act as a stand-alone summary for the entire document sets. Another drawback is that the clustering quality and cluster interpretation are treated as two isolated research problems and are studied separately. In this paper, we design and develop a unified system Gene Expression Miner to address these challenging issues in a principled and general manner by integrating cluster ensemble, text clustering, and multidocument summarization and provide an environment for comprehensive gene expression data analysis. We present a novel cluster ensemble approach to generate high-quality gene cluster. In our text summarization module, given a gene cluster, our expectation-maximization based algorithm can automatically identify subtopics and extract most probable terms for each topic. Then, the extracted top k topical terms from each subtopic are combined to form the biological explanation of each gene cluster. Experimental results demonstrate that our system can obtain high-quality clusters and provide informative key terms for the gene clusters.

We define and study a novel text mining problem for biomedical literature digital library, referred to as the class-attribute mining. Given a collection of biomedical literature from a digital library addressing a set of objects (e.g., proteins) and their descriptions (e.g., protein functions), the tasks of class-attribute mining include: (1) to identify and summarize latent classes in the space of objects, (2) to discover latent attribute themes in the space of object descriptions, and (3) to summarize the commonalities and differences among identified classes along each attribute theme. We approach this mining problem through a mixture language model and estimate the parameters of the model using the EM algorithm. We demonstrate the effectiveness of the model with an application called protein community identification and annotation from Medline, the largest biomedical literature digital library with more than 16 millions abstracts.

Document representation is one of the crucial components that determine the effectiveness of text classification tasks. Traditional document representation approaches typically adopt a popular bag-of-word method as the underlying document representation. Although it’s a simple and efficient method, the major shortcoming of bag-of-word representation is in the independent of word feature assumption. Many researchers have attempted to address this issue by incorporating semantic information into document representation. In this paper, we study the effect of semantic representation on the effectiveness of text classification systems. We employed a novel semantic smoothing technique to derive semantic information in a form of mapping probability between topic signatures and single-word features. Two classifiers, Naïve Bayes and Support Vector Machine, were selected to carry out the classification experiments. Overall, our topic-signature semantic representation approaches significantly outperformed traditional bag-of-word representation in most datasets.

Bayesian text classifiers face a common issue which is referred to as data sparsity problem, especially when the size of training data is very small. The frequently used Laplacian smoothing and corpus-based background smoothing are not effective in handling it. Instead, we propose a novel semantic smoothing method to address the sparse problem. Our method extracts explicit topic signatures (e.g. words, multiword phrases, and ontology-based concepts) from a document and then statistically maps them into single-word features. We conduct comprehensive experiments on three testing collections (OHSUMED, LATimes, and 20NG) to compare semantic smoothing with other approaches. When the size of training documents is small, the bayesian classifier with semantic smoothing not only outperforms the classifiers with background smoothing and Laplacian smoothing, but also beats the state-of-the-art active learning classifiers and SVM classifiers. In this paper, we also compare three types of topic signatures with respect to their effectiveness and efficiency for semantic smoothing.

With a growing number of works utilizing link information in enhancing document clustering, it becomes necessary to make a comparative evaluation of the impacts of different link types on document clustering. Various types of links between text documents, including explicit links such as citation links and hyperlinks, implicit links such as co-authorship links, and pseudo links such as content similarity links, convey topic similarity or topic transferring patterns, which is very useful for document clustering. In this study, we adopt a Relaxation Labeling (RL)-based clustering algorithm, which employs both content and linkage information, to evaluate the effectiveness of the aforementioned types of links for document clustering on eight datasets. The experimental results show that linkage is quite effective in improving content-based document clustering. Furthermore, a series of interesting findings regarding the impacts of different link types on document clustering are discovered through our experiments.

BACKGROUND: Affymetrix high density oligonucleotide expression arrays are widely used across all fields of biological research for measuring genome-wide gene expression. An important step in processing oligonucleotide microarray data is to produce a single value for the gene expression level of an RNA transcript using one of a growing number of statistical methods. The challenge for the researcher is to decide on the most appropriate method to use to address a specific biological question with a given dataset. Although several research efforts have focused on assessing performance of a few methods in evaluating gene expression from RNA hybridization experiments with different datasets, the relative merits of the methods currently available in the literature for evaluating genome-wide gene expression from Affymetrix microarray data collected from real biological experiments remain actively debated. RESULTS: The present study reports a comprehensive survey of the performance of all seven commonly used methods in evaluating genome-wide gene expression from a well-designed experiment using Affymetrix microarrays. The experiment profiled eight genetically divergent barley cultivars each with three biological replicates. The dataset so obtained confers a balanced and idealized structure for the present analysis. The methods were evaluated on their sensitivity for detecting differentially expressed genes, reproducibility of expression values across replicates, and consistency in calling differentially expressed genes. The number of genes detected as differentially expressed among methods differed by a factor of two or more at a given false discovery rate (FDR) level. Moreover, we propose the use of genes containing single feature polymorphisms (SFPs) as an empirical test for comparison among methods for the ability to detect true differential gene expression on the basis that SFPs largely correspond to cis-acting expression regulators. The PDNN method demonstrated superiority over all other methods in every comparison, whilst the default Affymetrix MAS5.0 method was clearly inferior. CONCLUSION: A comprehensive assessment of seven commonly used data extraction methods based on an extensive barley Affymetrix gene expression dataset has shown that the PDNN method has superior performance for the detection of differentially expressed genes.

Finding the right software is often hindered by different criteria as well as by technology changes. We performed an analytic hierarchy process (AHP) analysis using Expert Choice to determine which data mining package was best suitable for us. Deliberating a dozen alternatives and objectives led us to a series of pair-wise comparisons. When further synthesizing the results, Expert Choice helped us provide a clear rationale for the decision. The issue is that data mining technology is changing very rapidly. Our article focused only on the major suppliers typically available in the market place. The method and the process that we have used can be easily applied to analyze and compare other data mining software or knowledge management initiatives.

A typical microarray gene expression dataset is usually both extremely sparse and imbalanced. To select multiple highly informative gene subsets for cancer classification and diagnosis, a new Fuzzy Granular Support Vector Machine---Recursive Feature Elimination algorithm (FGSVM-RFE) is designed in this paper. As a hybrid algorithm of statistical learning, fuzzy clustering, and granular computing, the FGSVM-RFE separately eliminates irrelevant, redundant, or noisy genes in different granules at different stages and selects highly informative genes with potentially different biological functions in balance. Empirical studies on three public datasets demonstrate that the FGSVM-RFE outperforms state-of-the-art approaches. Moreover, the FGSVM-RFE can extract multiple gene subsets on each of which a classifier can be modeled with 100% accuracy. Specifically, the independent testing accuracy for the prostate cancer dataset is significantly improved. The previous best result is 86% with 16 genes and our best result is 100% with only eight genes. The identified genes are annotated by Onto-Express to be biologically meaningful.

Despite an influx of molecular data in the form of sequences, structure, transcription profiles etc., most of the protein interaction information relevant to cell biology research still exists strictly in the scientific literature which is written in a natural language that computers cannot easily manipulate. Automatically mining and extracting information from biomedical text holds the promise of easily consolidating large amounts of biological knowledge in computer-accessible form. In this talk, we present a novel approach Bio-IEDM (Biomedical Information Extraction and Data Mining) to integrate text mining and predictive modeling to analyze biomolecular network from biomedical literature databases. Our method consists of two phases. In phase 1, we discuss a semi-supervised efficient learning approach to automatically extract biological relationships such as protein-protein interaction, protein-gene interaction from the biomedical literature databases to construct the biomolecular network. In phase 2, we present a novel clustering algorithm to analyze the biomolecular network graph to identify biologically meaningful subnetworks (communities). The clustering algorithm considers the characteristics of the scale-free network graphs and is based on the local density of the vertex and its neighborhood functions that can be used to find more meaningful clusters with different density level. The experimental results indicate our approach is very effective in extracting biological knowledge from a huge collection of biomedical literatures. The integration of data mining and information extraction provides a promising direction for analyzing the biomolecular network.

Recent work has shown that ontology is useful to improve the performance of information retrieval, especially in biomedical literatures. The method of ontology-based can solve synonym problems. In this paper, we propose a new frame for genomic information retrieval based on UMLS. In our frame, genomic information retrieval includes three processes: first, documents were indexed based UMLS, which means documents were represented by concepts, besides, the concept weight was re-calculated combined with similarity between concepts. Second, documents were clustered using fuzzy c-means method. At last cluster language model is utilized for information retrieval. Our method can solve partly synonymy and polysemy problems. The new method is evaluated on TREC 2004/05 genomics track collections. Experiments show that the retrieval performance is greatly improved by the new method compared with the basic language model.

Hidden markov model (HMM) is frequently used for Pinyin-to-Chinese conversion. But it only captures the dependency with the preceding character. Higher order markov models can bring higher accuracy, but are computationally unaffordable to average PC settings. We propose a segment-based hidden markov model (SHMM), which has the same magnitude of complexity as first-order HMM, but generates higher decoding accuracy. SHMM tells a word from a bigram connecting two words, and assigns a reasonable probability to words as a whole. It is more powerful than HMM to decode words containing over two characters. We conduct a comprehensive Pinyin-to-Chinese conversion evaluation on Lancaster corpus. The experiment shows the perfect sentence accuracy is improved from 34.7% (HMM) to 43.3% (SHMM). The one-error sentence accuracy is increased from 72.7% to 78.3%. Furthermore, SHMM can seamlessly integrate with pinyin typing correction, acronym pinyin input, user-defined words, and self-adaptive learning all of which are a must for a commercial Pinyin-to-Chinese conversion product in order to improve the efficiency of pinyin input.

The majority of text retrieval and mining techniques are still based on exact feature (e.g. words) matching and unable to incorporate text semantics. Many researchers believe that the extension with semantic knowledge could improve the results and various methods (most of them are heuristic) have been proposed to account for concept hierarchy, synonymy, and other semantic relationships. However, the results with such semantic extension have been mixed, ranging from slight improvements to decreases in effectiveness, mostly likely due to the lack of a formal framework. Instead, we propose a novel method to address the semantic extension within the framework of language modeling. Our method extracts explicit topic signatures from documents and then statistically maps them into single- word features. The incorporation of semantic knowledge then reduces to the smoothing of unigram language models using semantic knowledge. The dragon toolkit reflects our method and its effectiveness is demonstrated by three tasks, text retrieval, text classification, and text clustering.

BACKGROUND: Biomolecular networks dynamically respond to stimuli and implement cellular function. Understanding these dynamic changes is the key challenge for cell biologists. As biomolecular networks grow in size and complexity, the model of a biomolecular network must become more rigorous to keep track of all the components and their interactions. In general this presents the need for computer simulation to manipulate and understand the biomolecular network model. RESULTS: In this paper, we present a novel method to model the regulatory system which executes a cellular function and can be represented as a biomolecular network. Our method consists of two steps. First, a novel scale-free network clustering approach is applied to the large-scale biomolecular network to obtain various sub-networks. Second, a state-space model is generated for the sub-networks and simulated to predict their behavior in the cellular context. The modeling results represent hypotheses that are tested against high-throughput data sets (microarrays and/or genetic screens) for both the natural system and perturbations. Notably, the dynamic modeling component of this method depends on the automated network structure generation of the first component and the sub-network clustering, which are both essential to make the solution tractable. CONCLUSION: Experimental results on time series gene expression data for the human cell cycle indicate our approach is promising for sub-network mining and simulation from large-scale biomolecular network.

BACKGROUND: A huge amount of biomedical textual information has been produced and collected in MEDLINE for decades. In order to easily utilize biomedical information in the free text, document clustering and text summarization together are used as a solution for text information overload problem. In this paper, we introduce a coherent graph-based semantic clustering and summarization approach for biomedical literature. RESULTS: Our extensive experimental results show the approach shows 45% cluster quality improvement and 72% clustering reliability improvement, in terms of misclassification index, over Bisecting K-means as a leading document clustering approach. In addition, our approach provides concise but rich text summary in key concepts and sentences. CONCLUSION: Our coherent biomedical literature clustering and summarization approach that takes advantage of ontology-enriched graphical representations significantly improves the quality of document clusters and understandability of documents through summaries.

In this paper, we present a novel approach Bio-IEDM (Biomedical Information Extraction and Data Mining) to integrate text mining and predictive modeling to analyze biomolecular network from biomedical literature databases. Our method consists of two phases. In phase 1, we discuss a semisupervised efficient learning approach to automatically extract biological relationships such as protein-protein interaction, protein-gene interaction from the biomedical literature databases to construct the biomolecular network. Our method automatically learns the patterns based on a few user seed tuples and then extracts new tuples from the biomedical literature based on the discovered patterns. The derived biomolecular network forms a large scale-free network graph. In phase 2, we present a novel clustering algorithm to analyze the biomolecular network graph to identify biologically meaningful subnetworks (communities). The clustering algorithm considers the characteristics of the scale-free network graphs and is based on the local density of the vertex and its neighborhood functions that can be used to find more meaningful clusters with different density level. The experimental results indicate our approach is very effective in extracting biological knowledge from a huge collection of biomedical literature. The integration of data mining and information extraction provides a promising direction for analyzing the biomolecular network.

Understanding the biomolecular network implementing cellular function goes beyond the old dogma of "one gene: one function"; only through comprehensive system understanding can we predict the impact of genetic variation in the population, design effective disease therapeutics, and evaluate the potential side-effects of therapies. In this paper, we present a novel method to model the regulatory system that executes a cellular function, which can be represented as a biomolecular network. Our method consists of three steps. First, the biomolecular network is derived using data-mining approaches to extend the initial conceptual biomolecular network from the literature search, etc. Secondly, once the whole biomolecular network structure is complete, a novel scale-free network clustering approach is applied to obtain various subnetworks. Lastly, fuzzy rule based models are generated for the subnetworks and simulations are run to predict their behavior in the cellular context. The modeling results represent hypotheses that are tested against high-throughput data sets (microarrays and/or genetic screens) for both the natural system and perturbations. If computational results do not match experimental or previously published results, then new hypotheses are formed and they feed back into the data-mining and analyzing step to refine the biomolecular network for the next iteration. This is repeated until a good match between modeling and data is obtained. Notably, the dynamic modeling component of this method depends on the automated network structure generation of the first component and the subnetwork clustering, which are both essential to make the solution tractable. Experimental results on human gene interaction networks and gene expression time series data for the human cell cycle indicate that our approach is promising for subnetwork mining and simulation from large biomolecular networks, as it produces a better convergence between continuous modeling and experiments.

Semantic smoothing, which incorporates synonym and sense information into the language models, is effective and potentially significant to improve retrieval performance. Previously implemented semantic smoothing models such as the translation model have shown good experimental results. However, these models are unable to incorporate contextual information. To overcome this limitation, we propose a novel context-sensitive semantic smoothing method that decomposes a document into a set of weighted context-sensitive topic signatures and then maps those topic signatures into query terms. The language model with such a context- sensitive semantic smoothing is referred to as the topic signature language model. In detail, we implement two types of topic signatures, depending on whether ontology exists in the application domain. One is the ontology-based concept and the other is the multiword phrase. The mapping probabilities from each topic signature to individual terms are estimated through the EM algorithm. Document models based on topic signature mapping are then derived. The new smoothing method is evaluated on the TREC 2004/ 2005 Genomics Track with ontology-based concepts, as well as the TREC Ad Hoc Track (Disks 1, 2, and 3) with multiword phrases. Both experiments show significant improvements over the two-stage language model, as well as the language model with context- insensitive semantic smoothing.

This paper discusses an ontology based language modeling text mining approach to the annotation of protein community. Communities appear to play an important role in the functional properties of complex networks. Being able to annotate the identified the community structure in a biological network can help us to understand better the structure and dynamics of biological systems. Traditional method such as gene ontology (GO) provides information about the functionality of gene products, but they are not enough to annotate community as for only limited number of proteins in the database, limited protein properties available for annotation and the inability to annotate a group of gene products as a whole. Thus, we present an ontology based mixture language model approach to annotate protein community. Compared to traditional method, we have the following three advantages. First, biomedical literature mining brings much richer information than existed gene databases. Second, the mixture language model can help "purify" the document by eliminating some background noise. Third, using domain ontology, we extract biological concept and concept pairs from abstracts. Biological concept is more meaningful than word or multi-word phrases. Moreover, using concept pairs can deliver much more information and serve as evidence of annotation results. We test our approach on four communities SAGA-SRB, CCR-NOT, RFC and ARP2/3, detected from dataset of interactions for Saccharomyces cerevisae from the general repository for interaction datasets (GRID). Annotation results provide a very coherent indication of functionality of each community

Document clustering has been used for better document retrieval, document browsing, and text mining. In this paper, we investigate if biomedical ontology MeSH improves the clustering quality for MEDLINE articles. For this investigation, we perform a comprehensive comparison study of various document clustering approaches such as hierarchical clustering methods (single-link, complete-link, and complete link), Bisecting K-means, K-means, and Suffix Tree Clustering (STC) in terms of efficiency, effectiveness, and scalability. According to our experiment results, biomedical ontology MeSH significantly enhances clustering quality on biomedical documents. In addition, our results show that decent document clustering approaches, such as Bisecting Kmeans, K-means and STC, gains some benefit from MeSH ontology while hierarchical algorithms showing the poorest clustering quality do not reap the benefit of MeSH ontology.

In this paper, we design and develop a unified system GE-Miner (Gene Expression Miner) to integrate cluster ensemble, text clustering and multi document summarization and provide an environment for comprehensive gene expression data analysis. We present a novel cluster ensemble approach to generate high quality gene cluster. In our text summarization module, given a gene cluster, our Expectation Maximization (EM) based algorithm can automatically identify subtopics and extract most probable terms for each topic. Then, the extracted top k topical terms from each subtopic are combined to form the biological explanation of each gene cluster. Experimental results demonstrate that our system can obtain high quality clusters and provide informative key terms for the gene clusters.

A document is often full of class-independent "general" words and short of class-specific "core " words, which leads to the difficulty of document clustering. We argue that both problems will be relieved after suitable smoothing of document models in agglomerative approaches and of cluster models in partitional approaches, and hence improve clustering quality. To the best of our knowledge, most model-based clustering approaches use Laplacian smoothing to prevent zero probability while most similarity-based approaches employ the heuristic TF*IDF scheme to discount the effect of "general" words. Inspired by a series of statistical translation language model for text retrieval, we propose in this paper a novel smoothing method referred to as context-sensitive semantic smoothing for document clustering purpose. The comparative experiment on three datasets shows that model-based clustering approaches with semantic smoothing is effective in improving cluster quality.

We introduce a novel document clustering approach that overcomes those problems by combining a semantic-based bipartite graph representation and a mutual refinement strategy. The primary contributions of this paper are the following. First, we introduce a new representation of documents using a bipartite graph between documents and co-occurrence concepts in the documents. Second, we show how to enhance clustering quality by applying the mutual refinement strategy to the initial clustering results. Third, through the experiments on MEDLINE documents, we show that our integrated method significantly enhances cluster quality and clustering reliability compared to existing clustering methods. Our approach improves on the average 29.5 cluster quality and 26.3 clustering reliability, in terms of misclassification index, over Bisecting K-means with the best parameters.

Two complementary and non-interactive literature sets of articles, when they are considered together, can reveal useful information of scientific interest not apparent in either of the two sets alone. Swanson called the existence of such hidden links as undiscovered public knowledge (UPK). The novel connection between Raynaud disease and fish oils was uncovered from complementary and non-interactive biomedical literature by Swanson in 1986. Since then, there have been many approaches to uncover UPK by mining the biomedical literature. These earlier works, however, required substantial manual intervention to reduce the number of possible connections. This paper proposes a semantic-based mining model for undiscovered public knowledge using the biomedical literature. Our method replaces manual ad-hoc pruning by using semantic knowledge from the biomedical ontologies. Using the semantic types and semantic relationships of the biomedical concepts, our prototype system can identify the relevant concepts collected from Medline and generate the novel hypothesis between these concepts. The system successfully replicates Swanson's two famous discoveries: Raynaud disease/fish oils and migraine/magnesium. Compared with previous approaches such as LSI-based and traditional association rule-based methods, our method generates much fewer but more relevant novel hypotheses, and requires much less human intervention in the discovery procedure.

Semantic smoothing, which incorporates synonym and sense information into the language models, is effective and potentially significant to improve retrieval performance. The implemented semantic smoothing models, such as the translation model which statistically maps document terms to query terms, and a number of works that have followed have shown good experimental results. However, these models are unable to incorporate contextual information. Thus, the resulting translation might be mixed and fairly general. To overcome this limitation, we propose a novel context-sensitive semantic smoothing method that decomposes a document or a query into a set of weighted context-sensitive topic signatures and then translate those topic signatures into query terms. In detail, we solve this problem through (1) choosing concept pairs as topic signatures and adopting an ontology-based approach to extract concept pairs; (2) estimating the translation model for each topic signature using the EM algorithm; and (3) expanding document and query models based on topic signature translations. The new smoothing method is evaluated on TREC 2004/05 Genomics Track collections and significant improvements are obtained. The MAP (mean average precision) achieves a 33.6 % maximal gain over the simple language model, as well as a 7.8 % gain over the language model with context-insensitive semantic smoothing.

In this paper we define the concepts of knowledge dependency, partial knowledge dependency and dependency degree in rough set models based on tolerance relations for incomplete information system. These concepts, not properly defined and studied in previous research study for incomplete information system, but are very important from data mining and rough set theory community. We consider two kinds of dependency relationships: one is related to the conditional attribute subset and the decision, and the other is related to two different conditional attribute subsets. We then propose several theorems based on our revised knowledge dependency relationships. We give formal proofs of the theorems and verify their correctness with some examples.

In this paper we present the design and evaluation of our biomedical literature searching approaches using the TREC 2004 ad hoc retrieval task in the genomics track. The main approach taken in our system is to expand queries by exploiting the three widely used strategies -local analysis, global analysis, and ontology-based term re-weighting across various search engines. The experimental results show that (1) ontology-based term re-weighting provides the best results among the three query expansion strategies, (2) expanding the initial query with more precise ontology-based term enhances LSI based local analysis substantially, and (3) including context to term re-weighting and LSI further improves the precision. Experimental results also show that the ontology-based term re-weighting with LUCENE or LEMUR search engines increases the average precision by up to 20.3% or 12.1%, respectively, compared to that of the baseline runs. In addition, the LSI-based local analysis increases the average precision by 9.2% with LEMUR search engine. We believe the approaches of the term re-weighting and LSI-based local analysis may be exploited in other bio-medical domains.

Selecting the most informative cancer-related genes from huge microarray gene expression data is an important and challenging bioinformatics research topic. This paper presents the novel granular support vector machines recursive feature elimination (GSVM-RFE) algorithm for the gene selection task. As a biologically meaningful hybrid method of statistical learning theory and granular computing theory, GSVM-RFE can separately eliminate irrelevant, redundant or noisy genes in different granules at different stages and can select positively related genes and negatively related genes in balance. Simulation results on the prostate cancer dataset show that GSVM-RFE is statistically much more accurate than traditional algorithms for the prostate cancer classification. More importantly, GSVM-RFE extracts a compact "perfect" gene subset of 17 genes with 100% accuracy. To our best knowledge, this is the first time such a "perfect" gene subset is reported, which is expected to be helpful for prostate cancer study.

Two complementary and non-interactive literature sets of articles, when they are considered together, can reveal useful information of scientific interest not apparent in either of the two document sets. Swanson called the existence of such knowledge, undiscovered public knowledge (UDPK). This paper proposes a semantic-based mining model for UDPK. Our method replaces manual ad-hoc pruning with using semantic knowledge from the biomedical ontologies. Using the semantic types and semantic relationships of the biomedical concepts, our prototype system can identify the relevant concepts collected from Medline and generate the novel hypothesis between these concepts. The system successfully replicates Swanson's two famous discoveries: Raynaud disease/fish oils and migraine/magnesium. Compared with previous approaches, our methods generate much fewer but more relevant novel hypotheses, and require much less human intervention in the discovery procedure.

Generating high quality gene clusters and identifying the underlying biological mechanism of the gene cluster are the important goals of clustering gene expression analysis. To get high quality cluster results, most of the current approaches rely on choosing the best cluster algorithm whose design biases and assumptions meet the underlying distribution of the data set. There are two issues for this approach: (1) usually the underlying data distribution of the gene expression data sets is unknown, and (2) there are so many clustering algorithms available and it is very challenging to choose the proper one. To provide a textual summary of the gene clusters, the most explored approach is the extractive approach that essentially builds upon techniques borrowed from the information retrieval, in which the objective is to provide terms to be used for query expansion, and not to act as a stand alone summary for the entire document sets. Another drawback is that the clustering quality and cluster interpretation are treated as two isolated research problems and are studied separately. But cluster quality and cluster interpretation are closely related and must be addressed in a coherent and unified way. It is essential to have relatively high quality clusters first, in order to get a correct, informative biological explanation of the gene cluster, otherwise, the biological explanation will be incorrect or misleading, no matter how good or robust the text summarization technique is. Based on this consideration, we design and develop a unified system GE-Miner (gene expression miner) to address these challenging issues in a principled and general manner by integrating cluster ensemble and text summarization and provide an environment for comprehensive gene expression data analysis. Experimental results demonstrate that our system can obtain high quality clusters and provide concise and informative textual summary for the gene clusters.

We present a biomedical literature data mining system SPIE-DM (Scalable and Portable Information Extraction and Data Mining) to extract and mine the protein-protein interaction network from biomedical literature such as MedLine. SPIE-DM consists of two phases: in phase 1, we develop a scalable and portable ie method (SPIE) to extract the protein-protein interaction from the biomedical literature. These extracted protein-protein interactions form a scale-free network graph. In phase 2, we apply a novel clustering method SFCluster to mine the protein-protein interaction network. The clusters in the network graph represent some potential protein complexes, which are very important for biologist to study the protein functionality. The clustering algorithm considers the characteristics of the scale-free network graphs and is based on the local density of the vertex and its neighborhood functions that can be used to find more meaningful clusters at different density levels. The experiments of SPIE-DM on around 1600 chromatin proteins indicate that our system is very promising for extracting and mining from biomedical literature databases.

Current search engines maintain a local repository to improve the search efficiency. A crawler is used to periodically poll the remote web pages to update the contents of the local repository. Due to the resource limitations, some local pages may be stale. To maintain the high freshness of the repository, the crawler is expected to revisit remote web pages in optimized order and frequency. The intuitive metric of freshness of the local repository is defined as the fraction of up-to-date web pages in the repository, which is merely based on the repository content, and does not, unfortunately, reflect the perspective of the search engine users, e.g., how often is a web page queried? We propose a novel weighted metric of the repository freshness with the importance of web pages being the weights. This metric not only takes into account the local web pages themselves but also the perspectives of the search engine users. We study the repository synchronization policy under this new metric, compare this metric with others, analyze its features, and discuss how the web page importance is determined.

Automated discovery and extraction of biological knowledge from biomedical web documents has become essential because of the enormous amount of biomedical literature published each year. In this paper we present an ontology-based scalable and portable information extraction system to automatically extract biological knowledge from huge collection of online biomedical web documents. Our method integrates ontology-based semantic tagging, information extraction and data mining together, automatically learns the patterns based on a few user seed tuples, and then extract new tuples from the biomedical web documents based on the discovered patterns. A novel system SPIE (Scalable and Portable Information Extraction) is implemented and tested on the PuBMed to find the chromatin protein-protein interaction and the experimental results indicate our approach is very effective in extracting biological knowledge from huge collection of biomedical web documents.

Web usage mining is the application of data mining techniques to discover usage patterns and behaviors from web data (clickstream, purchase information, customer information, etc.) in order to understand and serve e-commerce customers better and improve the online business. In this article, we present a general data warehouse/online analytic processing (OLAP) framework for web usage mining and business intelligence reporting. When we integrate the web data warehouse construction, data mining, and OLAP into the e-commerce system, this tight integration dramatically reduces the time and effort for web usage mining, business intelligence reporting, and mining deployment. Our data warehouse/OLAP framework consists of four phases: data capture, webhouse construction (clickstream marts), pattern discovery and cube construction, and pattern evaluation and deployment. We discuss data transformation operations for web usage mining and business reporting in clickstream, session, and customer levels; describe the problems and challenging issues in each phase in detail; provide plausible solutions to the issues; and demonstrate the framework with some examples from some real web sites. Our data warehouse/OLAP framework has been integrated into some commercial e-commerce systems. We believe this data warehouse/OLAP framework would be very useful for developing any real-world web usage mining and business intelligence reporting systems. © 2004 Wiley Periodicals, Inc.

In this paper we present a novel method to discover cyber communities from the WWW. A cyber community is a set of highly connected web pages in the WWW, which share the similar topics or interests. The WWW can be modeled as a huge scale-free network graph. Discovering cyber communities is converted to trawling the corresponding web graph to search for approximate p-quasi complete graph. The algorithm presented in this paper considers the characteristics of the scale-free network graphs and is based on the neighborhood information of the vertex in the p-quasi complete graph in the web graph.

Mining association rules from databases is a time-consuming process. Finding the large item set fast is the crucial step in the association rule algorithm. In this paper we present a fast association rule algorithm (Bit-AssoRule) based on granular computing. Our Bit-AssocRule doesn't follow the generation-and-test strategy of Apriori algorithm and adopts the divide-and-conquer strategy, thus avoids the time-consuming table scan to rind and prune the itemsets, all the operations of finding large itemsets from the datasets are the fast bit operations based on its corresponding granular. The experimental result of our Bit-AssocRule algorithm with Apriori, AprioriTid and AprioirHybrid algorithms shows Bit-AssocRule is 2 to 3 orders of magnitudes faster. Our research indicates that bitmap and granular computing can greatly improve the performance of association rule algorithm, and are very promising for data mining applications.

In this paper, we discuss some new bitmap techniques for optimizing decision support queries and association rule algorithm. We first show how to use a new type of predefined bitmap join index (prejoin/spl I.bar/bitmap/spl I.bar/index) to efficiently execute complex decision support queries with multiple outer join operations involved and push the outer join operations from the data flow level to the bitmap level and achieve significant performance gain. Then we discuss a bitmap based association rule algorithm. Our bitmap based association rule algorithm Bit-AssocRule doesn't follow the generation-and-test strategy of a priori algorithm and adopts the divide-and-conquer strategy, thus avoids the time-consuming table scan to find and prune the item sets, all the operations of finding large item sets from the datasets are the fast bit operations. The experimental results show Bit-AssocRule is 2 to 3 orders of magnitude faster than a priori and a priori hybrid algorithms. Our results indicate that bitmap techniques can greatly improve the performance of decision support queries and association rule algorithm, and bitmap techniques are very promising for the decision support query optimization and data mining applications.

Web usage mining describes the procedure, which mines Web data (clickstream, purchase information, etc.) to determine usage patterns. Web usage mining is needed in order to understand and serve e-commerce customers better. We present a general OLAM (on-line analytical mining) framework for Web usage mining and business intelligence reporting. Our OLAM framework consists of four phases: data capture, Webhouse construction (clickstream marts), pattern discovery and pattern evaluation. We describe the problems and challenging issues in each phase in detail and provide a general approach and guidelines to Web usage mining and business intelligence reporting for e-commerce.

The article presents a novel approach to constructing a good ensemble of classifiers using rough set theory and database operations. Ensembles of classifiers are formulated precisely within the framework of rough set theory and constructed very efficiently by using set-oriented database operations. Our method first computes a set of reducts which include all the indispensable attributes required for the decision categories. For each reduct, a reduct table is generated by removing those attributes which are not in the reduct. Next, a novel rule induction algorithm is used to compute the maximal generalized rules for each reduct table and a set of reduct classifiers is formed based on the corresponding reducts. The distinctive features of our method as compared to other methods of constructing ensembles of classifiers are: (1) presents a theoretical model to explain the mechanism of constructing ensemble of classifiers; (2) each reduct is a minimum subset of attributes and has the same classification ability as the entire attributes; (3) each reduct classifier constructed from the corresponding reduct has a minimal set of classification rules, and is as accurate and complete as possible and at the same time as diverse as possible from the other classifiers; (4) the test indicates that the number of classifiers used to improve the accuracy is much less than other methods.

We present a method to learn maximal generalized decision rules from databases by integrating discretization, generalization and rough set feature selection. Our method reduces the data horizontally and vertically. In the first phase, discretization and generalization are integrated and the numeric attributes are discretized into a few intervals. The primitive values of symbolic attributes are replaced by high level concepts and some obvious superfluous or irrelevant symbolic attributes are also eliminated. Horizontal reduction is accomplished by merging identical tuples after the substitution of an attribute value by its higher level value in a pre‐defined concept hierarchy for symbolic attributes, or the discretization of continuous (or numeric) attributes. This phase greatly decreases the number of tuples in the database. In the second phase, a novel context‐sensitive feature merit measure is used to rank the features, a subset of relevant attributes is chosen based on rough set theory and the merit values of the features. A reduced table is obtained by removing those attributes which are not in the relevant attributes subset and the data set is further reduced vertically without destroying the interdependence relationships between classes and the attributes. Then rough set‐based value reduction is further performed on the reduced table and all redundant condition values are dropped. Finally, tuples in the reduced table are transformed into a set of maximal generalized decision rules. The experimental results on UCI data sets and a real market database demonstrate that our method can dramatically reduce the feature space and improve learning accuracy.

We present a method to mine maximal generalized decision rules from databases by integrating discretization, generalization and rough sets feature selection. Our method reduces the data horizontally and vertically. In the first phase, discretization and generalization are integrated and the numeric attributes are discretized into a few intervals. Primitive values of symbolic attributes are replaced by high level concepts and some obvious superfluous or irrelevant symbolic attributes are also eliminated. Horizontal reduction is accomplished by merging identical tuples after the substitution of an attribute value by its higher level value in a predefined concept hierarchy for symbolic attributes or the discretization of continuous (or numeric) attributes. In the second phase, a novel context sensitive feature merit measure is used to rank the features, a subset of relevant attributes is chosen based on rough sets theory and the merit values of the features. A reduced table is obtained by removing those attributes which are not in the relevant attributes subset and the data set is further reduced vertically without destroying the interdependence relationships between the classes and the attributes. Rough sets based value reduction is further performed on the reduced table and all redundant condition values are dropped, finally, tuples in the reduced table are transformed into a set of maximal generalized decision rules. The experimental results on UCI data sets and an actual market database shows that our method can dramatically reduce the feature space and improve the learning accuracy.

The principle and experimental results of an attribute oriented rough set approach for knowledge discovery in databases are described. Our method integrates the database operation, rough set theory and machine learning techniques. In this method the learning procedure consists of two phases: data generalization and data reduction. In the data generalization phase, attribute oriented induction is performed attribute by attribute using attribute removal and concept ascension, some undesirable attributes to the discovery task are removed and the primitive data is generalized to the desirable level; thus a set of tuples may be generalized to the same generalized tuple. This procedure substantially reduces the computational complexity of the database learning process. Subsequently, in data reduction phase, the rough set method is applied to the generalized relation to find a minimal attribute set relevant to the learning task. The generalized relation is reduced further by removing those attributes which are irrelevant and/or unimportant to the learning task. Finally the tuples in the reduced relation are transformed into different knowledge rules based on different knowledge discovery algorithms. Based upon these principles, a prototype knowledge discovery system, DBROUGH has been constructed. In DBROUGH, a variety of knowledge discovery algorithms are incorporated and different kinds of knowledge rules, such as characteristic rules, classification rules, decision rules, maximal generalized rules can be discovered efficiently and effectively from large databases.

Knowledge discovery in databases, or dala mining, is an important direction in the development of data and knowledge‐based systems. Because of the huge amount of data stored in large numbers of existing databases, and because the amount of data generated in electronic forms is growing rapidly, it is necessary to develop efficient methods to extract knowledge from databases. An attribute‐oriented rough set approach has been developed for knowledge discovery in databases. The method integrates machine‐learning paradigm, especially learning‐from‐examples techniques, with rough set techniques. An attribute‐oriented concept tree ascension technique is first applied in generalization, which substantially reduces the computational complexity of database learning processes. Then the cause‐effect relationship among the attributes in the database is analyzed using rough set techniques, and the unimportant or irrelevant attributes are eliminated. Thus concise and strong rules with little or no redundant information can be learned efficiently. Our study shows that attribute‐oriented induction combined with rough set theory provide an efficient and effective mechanism for knowledge discovery in database systems.

We develop an attribute-oriented rough set approach for the discovery of decision rules in relational databases. Our approach combines machine learning techniques and rough set theory. We consider a learning procedure to consist of the two phases data generalization and data reduction. In the data generalization phase, utilizing knowledge about concept hierarchies and relevance of the data, an attribute-oriented induction is performed attribute by attribute. Some undesirable attributes of the discovery task are removed and the primitive data in the databases are generalized to the desirable level; this process greatly decreases the number of tuples which must be examined for the discovery task and substantially reduces the computational complexity of the database learning processes. Subsequently, in data reduction phase, rough set theory is applied to the generalized relation; the cause-effect relationships among the condition and decision attributes in the databases are analyzed and the non-essential or irrelevant attributes to the discovery task are eliminated without losing information of the original database system. This process further reduces the generalized relation. Thus very concise and more accurate decision rules for each class in the decision attribute with little or no redundancy information, can be extracted automatically from the reduced relation during the learning process. Our study shows that attribute-oriented induction combined with rough set theory provide an efficient and effective mechanism for discovering decision rules in database systems.