MOTIVATION: Cryo-electron microscopy (Cryo-EM) single particle analysis (SPA) is a key technique for revealing the structure of biomacromolecules by three-dimensional reconstruction. Achieving high-resolution reconstruction relies on the acquisition of a large number of authentic particles; however, manual particle picking is inefficient and inadequate for the demands of reconstruction, making automated particle picking a major research focus. Although the foundational segmentation model Segment Anything Model (SAM) has recently advanced automated particle picking, its segmentation advantages have not been fully realized in cryo-EM applications. Moreover, cryo-EM images often have significant noise. Conventional denoising decreases noise but frequently overlooks high-level semantic information, leading to oversmoothed particle regions and reduced particle distinguishability. RESULTS: To address these challenges, we propose CryoPromptSeg, which employs prompt-guided SAM for particle picking while integrating a semantically enhanced image denoiser. Specifically, by performing domain adaptation fine-tuning of SAM and incorporating prompts generated by the proposed automatic prompt generator, it achieves precise segmentation of cryo-EM particles. In addition, it employs a parallel multi-task framework to jointly train the denoiser and the prompt generator, incorporating particle semantic information from the prompt generator into the denoiser to suppress noise while preserving highly distinguishable particle structures. To lower the barrier to practical application, we developed a user-friendly online prediction platform for particle picking. Experimental results demonstrate that CryoPromptSeg outperforms existing mainstream methods in both particle picking accuracy and image denoising quality, thus providing a novel solution for the automation of particle picking. AVAILABILITY: The code and platform are available at: https://github.com/347251369/CryoPromptSeg.

With the continuous advancement of Internet technology, public demand for healthcare services has been growing rapidly. However, the overall distribution of medical resources remains significantly unbalanced. To promote the development of hierarchical diagnosis and treatment systems, the construction of an Internet hospital platform that enables seamless interconnection and resource sharing among all participants in the healthcare process is of critical importance to deliver efficient and high-quality medical services. Currently, existing Internet hospital platforms face challenges such as limited scalability and inefficient medical data circulation. To address these problems, this study proposes a three-tier architecture for Internet hospital platforms by designing a unified overall structure. Based on this architecture, we develop a data exchange platform that can effectively meet the needs of various participants in the medical system. Application results demonstrate that the platform offers high scalability and effectively enhances the efficiency of medical resource circulation.

MOTIVATION: Studies on pan-cancer related genes play important roles in cancer research and precision therapy. With the richness of research data and the development of neural networks, several successful methods that take advantage of multiomics data, protein interaction networks, and graph neural networks to predict cancer genes have emerged. However, these methods also have several problems, such as ignoring potentially useful biological data and providing limited representations of higher-order information. RESULTS: In this work, we propose a pan-cancer related gene predictive model, the DGHNN, which takes biological pathways into consideration, applies a deep graph and hypergraph neural network to encode the higher-order information in the protein interaction network and biological pathway, introduces skip residual connections into the deep graph and hypergraph neural network to avoid problems with training the deep neural network, and finally uses a feature tokenizer and transformer for classification. The experimental results show that the DGHNN outperforms other methods and achieves state-of-the-art model performance for pan-cancer related gene prediction. AVAILABILITY AND IMPLEMENTATION: The DGHNN is available at https://github.com/skytea/DGHNN.

The recent advent of in-context learning (ICL) capabilities in large pre-trained models has yielded significant advancements in the generalization of segmentation models. By supplying domain-specific image-mask pairs, the ICL model can be effectively guided to produce optimal segmentation outcomes, eliminating the necessity for model fine-tuning or interactive prompting. However, current existing ICL-based segmentation models exhibit significant limitations when applied to medical segmentation datasets with substantial diversity. To address this issue, we propose a dual similarity checkup approach to guarantee the effectiveness of selected in-context samples so that their guidance can be maximally leveraged during inference. We first employ large pre-trained vision models for extracting strong semantic representations from input images and constructing a feature embedding memory bank for semantic similarity checkup during inference. Assuring the similarity in the input semantic space, we then minimize the discrepancy in the mask appearance distribution between the support set and the estimated mask appearance prior through similarity-weighted sampling and augmentation. We validate our proposed dual similarity checkup approach on eight publicly available medical segmentation datasets, and extensive experimental results demonstrate that our proposed method significantly improves the performance metrics of existing ICL-based segmentation models, particularly when applied to medical image datasets characterized by substantial diversity.

Previous research has identified hematoma as the primary cause of mortality and disability in intraventricular hemorrhage (IVH), yet neurologic outcomes post-IVH have not significantly improved following hematoma evacuation. Therefore, utilizing spatial transcriptomics sequencing technology to acquire and analyze IVH data is of significant importance for the early diagnosis and treatment of acute cerebral hemorrhage. However, existing spatial transcriptomics clustering algorithms often overlook local information, fail to integrate multimodal data, and suffer from performance instability. To address these issues, we developed a multimodal spatial transcriptomics clustering algorithm named STARGATE, which initially captures the global spatial topology and local long-sequence gene set features of cells by integrating Graph Convolutional Networks (GCNs) based on self-attention mechanisms and Long Short-Term Memory networks (LSTMs), subsequently fuses gene expression, tissue imagery, morphological, and spatial coordinate features using Gated Recurrent Units (GRUs), and finally enhances the accuracy and robustness of predictions through ensemble learning methods. Our key experimental results include: (1) In clustering experiments on our proprietary mouse IVH spatial transcriptomics dataset, the STARGATE model achieved Adjusted Rand Index (ARI), Normalized Mutual Information (NMI), and Fowlkes-Mallows Index (FMI) scores of 0.80, 0.84, and 0.82, respectively, all superior to classic algorithms such as stLearn and CCST. (2) In clustering experiments on the mouse IVH dataset, the STARGATE model reached an accuracy of 89.63%, with an excellent confusion matrix, demonstrating outstanding accuracy and stability. (3) In clustering experiments on the public dataset DLFPC, which is known for its high data complexity leading to generally lower clustering metrics, the STARGATE model's ARI, NMI, and FMI indicators still outperformed classic algorithms such as SpaGCN and Seurat.

The continuous mutation and evolution of H5N1 may cause a wide spread of the epidemic, however, the mechanism of H5N1 mutation and evolution are still unclear. Therefore, further mining the overall evolutionary pattern of H5N1 sequences, studying the impact of global factors such as migratory bird flyways on H5N1 evolution, and constructing a corresponding web service are crucial for preventing potential future H5N1 epidemics. To this end, we firstly analyzed the within-host mutation features of H5N1 sequences and optimized the phylogenetic model. Secondly, we employed the optimized model to construct the phylogenetic trees and performed H5N1 evolution group analysis. Thirdly, we not only analyzed the correlation between different evolution groups of H5N1 and global migratory bird flyways, but also investigated the impact of the El Niño climate phenomenon on the evolution of H5N1. Finally, we built up a web service for H5N1 mutation and evolution analysis and visualization. Our main results include: (1) We found that the global H5N1 can be categorized into three major evolution groups at different time in various geographic regions. (2) We not only found that all three major groups of HA fragments were related to the East Asian -Australasian flyway of migratory birds, but also demonstrated the correlation and causality between the Oceanic Niño Index and the alternation of old and new evolutionary groups of H5N1. (3) VP-H5N1 provides a fast and easy-to-use web service platform for online analysis and visualization of H5N1 evolution.

Vaccination stands as the most effective and economical strategy for prevention and control of influenza. The primary target of neutralizing antibodies is the surface antigen hemagglutinin (HA). However, ongoing mutations in the HA sequence result in antigenic drift. The success of a vaccine is contingent on its antigenic congruence with circulating strains. Thus, predicting antigenic variants and deducing antigenic clusters of influenza viruses are pivotal for recommendation of vaccine strains. The antigenicity of influenza A viruses is determined by the interplay of amino acids in the HA1 sequence. In this study, we exploit the ability of convolutional neural networks (CNNs) to extract spatial feature representations in the convolutional layers, which can discern interactions between amino acid sites. We introduce PREDAC-CNN, a model designed to track antigenic evolution of seasonal influenza A viruses. Accessible at http://predac-cnn.cloudna.cn, PREDAC-CNN formulates a spatially oriented representation of the HA1 sequence, optimized for the convolutional framework. It effectively probes interactions among amino acid sites in the HA1 sequence. Also, PREDAC-CNN focuses exclusively on physicochemical attributes crucial for the antigenicity of influenza viruses, thereby eliminating unnecessary amino acid embeddings. Together, PREDAC-CNN is adept at capturing interactions of amino acid sites within the HA1 sequence and examining the collective impact of point mutations on antigenic variation. Through 5-fold cross-validation and retrospective testing, PREDAC-CNN has shown superior performance in predicting antigenic variants compared to its counterparts. Additionally, PREDAC-CNN has been instrumental in identifying predominant antigenic clusters for A/H3N2 (1968-2023) and A/H1N1 (1977-2023) viruses, significantly aiding in vaccine strain recommendation.

Peptide binding to major histocompatibility complex (MHC) proteins plays a critical role in T-cell recognition and the specificity of the immune response. Experimental validation such peptides is extremely resource-intensive. As a result, accurate computational prediction of binding peptides is highly important, particularly in the context of cancer immunotherapy applications, such as the identification of neoantigens. In recent years, there is a significant need to continually improve the existing prediction methods to meet the demands of this field. We developed ConvNeXt-MHC, a method for predicting MHC-I-peptide binding affinity. It introduces a degenerate encoding approach to enhance well-established panspecific methods and integrates transfer learning and semi-supervised learning methods into the cutting-edge deep learning framework ConvNeXt. Comprehensive benchmark results demonstrate that ConvNeXt-MHC outperforms state-of-the-art methods in terms of accuracy. We expect that ConvNeXt-MHC will help us foster new discoveries in the field of immunoinformatics in the distant future. We constructed a user-friendly website at http://www.combio-lezhang.online/predict/, where users can access our data and application.

It is important for mental health research to study the scale of the elementary and secondary school students. Methods: This study uses knowledge graph technology to strengthen the relationship between individual psychological questionnaires and various test indicators of mental health, and we develop the lightweight extraction method "Relational database to Knowledge graph" (RETG) to generate the necessary files to build the knowledge graph. We not only employ quantum search algorithm to reduce the computational complexity, but also replace multi bit quantum gate in the quantum circuit by combining single- and two-bit quantum gate. Results: We built a scalable adaptive mental health testing platform for the elementary and secondary school students; The RETG method increases the efficiency of Knowledge graph construction; The quantum computing can reduce the time complexity of construction process and we increase the accuracy of quantum acceleration algorithms on physical machines.

It is very potential to develop digital villages for promoting smart agriculture. As one of the important research fields of smart agriculture, smart chicken farms encounter management problems such as difficulties in quickly and accurately warning of sick and dead chickens and estimating feed residuals. Therefore, this study not only respectively proposed CKTrack and FRCM to detect sick and dead chickens and estimate feed residuals, but also developed a smart chicken farming platform for automagical management. Our main results include (1) the proposed CKTrack method can effectively identify sick and dead chickens under the condition of limited data volume and computing capacity; (2) the proposed FRCM method can accurately estimate the feed residuals; and (3) the smart chicken farming platform developed can provide farmers with functions such as early warning of sick and dead chickens, visualization of the chicken quantity inventory, and feed residual estimation.

Radiomics is an important research direction in the field of medical image analysis. Although the number of publications is increasing year by year, it has been difficult to translate into clinical practice due to the small size of clinical data. In most cases, Radiomics data can be considered as small tabular data. Deep learning is often less effective than classical machine learning algorithms in processing tabular data. Recently, table representation learning has started to receive more widespread attention which is often an easily overlooked but very important area of research, helping improve the status of tabular deep learning. Here, we first apply a pre-trained Transformer model named Tabular Prior-Data Fitted Network (TabPFN) to the field of Radiomics analysis. We implement extensive experiments on three real-world clinical datasets: (a) Ultrasound Radiomics dataset for the classificatory diagnosis of Kidney tumor, (b) CT Radiomics dataset for the prediction of EGFR gene mutations in non-small cell lung cancer, (c) MRI Radiomics dataset for the prediction of treatment response of brain metastases to gamma knife radiosurgery. By comprehensive analysis, we demonstrate that the pre-trained tabular Transformer can be used as a realtime, efficient, and stable Radiomics data processor with superior performance over other tabular machine learning methods in different clinical tasks. We also simulate an ideal clinical practice scenario for evaluating the clinical translation potential of pretrained models. Finally, we explore the advantages and limitations of pre-trained tabular models for Radiomics analysis.

lass imbalance and data obsolescence are two major issues in the field of credit scoring, leading to excessive bias and inaccuracies in the classification process of credit scoring models. In order to augment minority class samples and balance time dependencies, this paper proposes a credit scoring model based on Temporal Sample Interaction Convolutional Network (TSICN) to facilitate better credit risk assessment for financial institutions. It relies on the intrinsic features of the minority class samples to synthesize new data, thereby increasing the quantity of the minority class samples. Through inference and synthesis, It greatly mitigates the loss of information from minority class samples. The synthesized minority class samples, blended with the original data, are inputted into the causal convolutional layers and dilated convolutional layers. The information flow and memory updates are regulated through reset gate and update gate, The reset gate determines how to combine past information with the current input, while the update gate determines how to combine the previous hidden state with the current candidate hidden state. Compared to common methods, TSICN can integrate information features from minority class samples into the synthesis data, focusing more on short-term dependencies while reducing the capture of long-term dependencies. Experimental results show that TSICN achieves excellent credit scoring classification performance on real-world datasets. This enables more accurate prediction of applicant credit risk, thus reducing the risk of loan default.

Advanced metering infrastructure (AMI) is a vital component of the smart grid (SG) for real-time data access and bidirectional communication. An authentication and key agreement (AKA) protocol is needed for AMI systems to ensure the confidentiality and integrity of communication data. Since the devices are connected to the open network and generally deployed outdoors with limited computation, communication, and storage, designing a suitable AKA protocol is a challenging task. Researchers are still looking for good ways to make the SG secure and efficient simultaneously. To remedy the situation, in this article, we advance a security-enhanced elliptic-curve-cryptography-based AKA protocol, and the security has been proven rigorously under the random oracle model and verified with the ProVerif tool. Furthermore, performance comparison validates the proposed protocol in affording improved security features with lower computation and communication cost. In addition, the proposed scheme is implemented practically on a testbed, which is deployed usingRaspberry Pi 3 Model B+for smart meters.

Ultrasound based estimation of fetal biometry is extensively used to diagnose prenatal abnormalities and to monitor fetal growth, for which accurate segmentation of the fetal anatomy is a crucial prerequisite. Although deep neural network-based models have achieved encouraging results on this task, inevitable distribution shifts in ultrasound images can still result in severe performance drop in real world deployment scenarios. In this article, we propose a complete ultrasound fetal examination system to deal with this troublesome problem by repairing and screening the anatomically implausible results. Our system consists of three main components: A routine segmentation network, a fetal anatomical key points guided repair network, and a shape-coding based selective screener. Guided by the anatomical key points, our repair network has stronger cross-domain repair capabilities, which can substantially improve the outputs of the segmentation network. By quantifying the distance between an arbitrary segmentation mask to its corresponding anatomical shape class, the proposed shape-coding based selective screener can then effectively reject the entire implausible results that cannot be fully repaired. Extensive experiments demonstrate that our proposed framework has strong anatomical guarantee and outperforms other methods in three different cross-domain scenarios.

With the development of the modern smart city, sharing-bikes require behaviors prediction for grid-level areas which is essential for intelligent transportation systems. A model which can predict bike sharing demand behaviours accurately can allocate sharing-bikes in advance to satisfy travel demands alongside saving energy, reducing traffic, cutting down waste for those sharing-bikes companies putting excessive sharing-bikes in unsaturated demand areas. In this paper, we abandon the traditional time series prediction method and use a more efficient deep learning method to solve the traffic forecasting problem. Moreover, instead of considering spatial relation and temporal relation relatively, we produced a deep multi-view spatial-temporal network to combine them into one prediction model framework. In the experimental section, we investigate in the experiment on enormous amount of real sharing-bikes application use data in the core region of Beijing to test the performance of the model framework with a 1 km$\times $1 km grid-level scale and compare it with other existing machine learning approaches and prediction models. And the 4G/5G/6G communication technology facilitate the real-time control of the space-time locations of sharing bikes dynamically. Thus, it provides the basis for high-frequency analysis of space-time patterns, especially supported by the 6G large-scale application in the future.

As an acute zoonosis, rabies has a fatality rate of nearly 100%. Since rabies surveillance data over the years show that the rabies epidemic areas in China expand from southern provinces to central and northern provinces, we selected Hubei Province, an area with a high incidence of rabies in Central China, to investigate national rabies prevention and control work by developing a visual analysis platform (VAP-HRHB: http://www.combio-lezhang.online/rabies/index.html). VAP-HRHB employs bioinformatics methods to predict the future developmental trend for rabies incidence and exposure numbers, thereby comprehensively improving the prevention and control of infectious diseases in China.

By reviewing previous CpG-related studies, we consider that the transcription regulation of about half of the human genes, mostly housekeeping (HK) genes, involves CpG islands (CGIs), their methylation states, CpG spacing and other chromosomal parameters. However, the precise CGI definition and positioning of CGIs within gene structures, as well as specific CGI-associated regulatory mechanisms, all remain to be explained at individual gene and gene-family levels, together with consideration of species and lineage specificity. Although previous studies have already classified CGIs into high-CpG (HCGI), intermediate-CpG (ICGI) and low-CpG (LCGI) densities based on CpG density variation, the correlation between CGI density and gene expression regulation, such as co-regulation of CGIs and TATA box on HK genes, remains to be elucidated. First, this study introduces such a problem-solving protocol for human-genome annotation, which is based on a combination of GTEx, JBLA and Gene Ontology (GO) analysis. Next, we discuss why CGI-associated genes are most likely regulated by HCGI and tend to be HK genes; the HCGI/TATA± and LCGI/TATA± combinations show different GO enrichment, whereas the ICGI/TATA± combination is less characteristic based on GO enrichment analysis. Finally, we demonstrate that Hadoop MapReduce-based MR-JBLA algorithm is more efficient than the original JBLA in k-mer counting and CGI-associated gene analysis.

MOTIVATION: The growth and survival of myeloma cells are greatly affected by their surrounding microenvironment. To understand the molecular mechanism and the impact of stiffness on the fate of myeloma-initiating cells (MICs), we develop a systems biological model to reveal the dynamic regulations by integrating reverse-phase protein array data and the stiffness-associated pathway. RESULTS: We not only develop a stiffness-associated signaling pathway to describe the dynamic regulations of the MICs, but also clearly identify three critical proteins governing the MIC proliferation and death, including FAK, mTORC1 and NFκB, which are validated to be related with multiple myeloma by our immunohistochemistry experiment, computation and manually reviewed evidences. Moreover, we demonstrate that the systematic model performs better than widely used parameter estimation algorithms for the complicated signaling pathway. AVAILABILITY AND IMPLEMENTATION: We can not only use the systems biological model to infer the stiffness-associated genetic signaling pathway and locate the critical proteins, but also investigate the important pathways, proteins or genes for other type of the cancer. Thus, it holds universal scientific significance. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Since the COVID-19 epidemic is still expanding around the world and poses a serious threat to human life and health, it is necessary for us to carry out epidemic transmission prediction, whole genome sequence analysis, and public psychological stress assessment for 2019-nCoV. However, transmission prediction models are insufficiently accurate and genome sequence characteristics are not clear, and it is difficult to dynamically assess the public psychological stress state under the 2019-nCoV epidemic. Therefore, this study develops a 2019nCoVAS web service (http://www.combio-lezhang.online/2019ncov/home.html) that not only offers online epidemic transmission prediction and lineage-associated underrepresented permutation (LAUP) analysis services to investigate the spreading trends and genome sequence characteristics, but also provides psychological stress assessments based on such an emotional dictionary that we built for 2019-nCoV. Finally, we discuss the shortcomings and further study of the 2019nCoVAS web service.

Since the expression and localization of protein plays the important role for GC, this study is to explore the connections between these proteomic information and GC prognosis. After we use bioinformatics method to analyze the dataset from Human Protein Atlas, we found that the proportions of high expression intensity in GC are less than those in normal stomach tissue (NST) samples, whereas the proportion of low expression intensity in GC are greater than those in NST; the proteins located in the nucleus of normal cells are more stable than others; MEN1 is such a tumor suppressor gene that has favorable prognosis.

Because of the different genetic structures and functional gene diversity, bacterial genome data have high complexity and dimensions. Therefore, it is difficult to reveal the sequence patterns and biological mechanism of a genome with classical analysis methods. Since artificial intelligence (AI) applications are capable of mining key biological information from massive multidimensional data, they are broadly employed to analyze bacterial genomes. However, to our knowledge, there are few systematic reviews that illustrate these AI applications in bacterial genomics research. Therefore, we first introduce the characteristics of bacterial genomics, and then briefly summarize AI applications in bacterial genomics research from three aspects: gene finding, gene function prediction and gene expression network construction. Finally, we discuss the challenges and future AI applications in bacterial genomics research.

BACKGROUND: Cervical cancer is the fourth most common tumor in women worldwide, mostly resulting from high-risk human papillomavirus (HR-HPV) with persistent infection. RESULTS: The present discoveries are comprised of the following: (i) A total of 16.64% of the individuals were positive for HR-HPV infection, with 13.04% having a single HR-HPV type and 3.60% having multiple HR-HPV types. (ii) Cluster analysis showed that the infection rate trends of HPV31 and HPV33 in all infections as well as HPV33 and HPV35 in single infections in precancerous stages were very similar. (iii) The single/multiple infection proportions of HR-HPV demonstrated a trend that the multiple infections rates of HR-HPV increased as the disease developed. CONCLUSIONS: The HR-HPV prevalence in outpatients was 16.64%, and the predominant HR-HPV types in the study were HPV52, HPV58 and HPV16. HR-HPV subtypes with common biological properties had similar infection rate trends in precancerous stages. Especially, as the disease development of precancer evolved, defense against HPV infection broke, meanwhile, the potential of more HPV infection increased, which resulted in increase of multiple infections of HPV.

BACKGROUND: Glioblastoma multiforme (GBM) is one of the most common malignant brain tumors and its average survival time is less than 1 year after diagnosis. RESULTS: Firstly, this study aims to develop the novel survival analysis algorithms to explore the key genes and proteins related to GBM. Then, we explore the significant correlation between AEBP1 upregulation and increased EGFR expression in primary glioma, and employ a glioma cell line LN229 to identify relevant proteins and molecular pathways through protein network analysis. Finally, we identify that AEBP1 exerts its tumor-promoting effects by mainly activating mTOR pathway in Glioma. CONCLUSIONS: We summarize the whole process of the experiment and discuss how to expand our experiment in the future.

It is well known that CpG island plays an important role in gene methylation. Since CpG island is closely related to human genetic characteristics such as TATA-box, tissue expression specificity, and LAUPs (Lineage-associated Underrepresented Permutations), it is important to investigate the sequence specificity of CpG island as well as the potential genetic characteristics related to CpG island to further understand the methylation related regulation mechanism. Therefore, this study develops such an online service website for CpG island related density and LAUPs analysis (CGIDLA, www.combio-lezhang.online/cgidla/index.html), that not only can investigate the relationship among the CpG island density, TATA-box feature, and expression breadth of human genes, but also deposit LAUPs of 32 representative species to help molecular biologists investigate the relationship between CpG island and LUAPs. Moreover, CGIDLA provides the source code download service and the related LAUPs counting functions.

The remarkable progress in single-cell RNA sequencing (scRNA-seq) in recent years has led to the heat of great development of computational methods and tools for scRNA-seq data analysis. They have seen wide applications in data preprocessing, discovering differential expression, cell grouping, and trajectory inference, etc. To summarize and compare these diverse methods and tools, this paper reviews the algorithms and software including the data preprocessing, cell grouping, and trajectory inference for conducting cell-level scRNA-seq studies.

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.

BACKGROUND: Haemorrhagic stroke accounts for approximately 31.52% of all stroke cases, and the most common origin is hypertension. However, little is known about the method to identify high-risk populations of hypertensive intracerebral haemorrhage. RESULTS: The results showed that the angle between the middle cerebral artery and the internal carotid artery (AMIC), the distance between the beginning of the median artery and superior trunk (DMS), and the density (CT value) of the lenticulostriate artery (CTL) were statistically significant enough to cause intracerebral haemorrhage. In addition, we chose these three potential features for the ensemble learning classification model. Our developed ensemble-learning method outperforms not only previous work but also three other classic classification methods based on accuracy measurements. CONCLUSIONS: The developed mathematical model in the present study is efficient in predicting the probability of intracerebral haemorrhage.

MOTIVATION: Recently, copy number variation (CNV) has gained considerable interest as a type of genomic variation that plays an important role in complex phenotypes and disease susceptibility. Since a number of CNV detection methods have recently been developed, it is necessary to help investigators choose suitable methods for CNV detection depending on their objectives. For this reason, this study compared ten commonly used CNV detection applications, including CNVnator, ReadDepth, RDXplorer, LUMPY and Control-FREEC, benchmarking the applications by sensitivity, specificity and computational demands. Taking the DGV gold standard variants as a standard dataset, we evaluated the ten applications with real sequencing data at sequencing depths from 5X to 50X. Among the ten methods benchmarked, LUMPY performs the best for both high sensitivity and specificity at each sequencing depth. For the purpose of high specificity, Canvas is also a good choice. If high sensitivity is preferred, CNVnator and RDXplorer are better choices. Additionally, CNVnator and GROM-RD perform well for low-depth sequencing data. Our results provide a comprehensive performance evaluation for these selected CNV detection methods and facilitate future development and improvement in CNV prediction methods.

If high sensitivity is preferred, CNVnator and RDXplorer are better choices.'Should be 'If high sensitivity is preferred, CNVnator and GROM-RD are better choices.'In Results: Sensitivity and Specificity of CNV prediction-first paragraph: 'At a low sequencing depth of 5X, the TPR of LUMPY reached 0.432, followed by CNVnator (0.370) and GROM-RD (0.359), which was much greater than other methods (0.021 to 0.254), implying that these three methods have greater sensitivity at low sequencing depth.At high sequencing depths of 30X and 50X, CNVnator also showed the highest TPR of 0.725 and 0.800, followed by LUMPY (0.711, 0.753) and RDXplorer (0.678, 0.621), implying higher sensitivity than other methods.Overall, at each sequencing depth from low to high, CNVnator and LUMPY had the best performance with respect to the sensitivity of CNV detection.'Should be 'At a low sequencing depth of 5X, the TPR of LUMPY reached 0.432, followed by GROM-RD (0.359) and CNVnator (0.287), which was much greater than other methods (0.021 to 0.254), implying that these three methods have greater sensitivity at low sequencing depth.At high sequencing depths of 30X and 50X, LUMPY also showed the highest TPR of 0.711 and 0.753, followed by CNVnator (0.510 and 0.542) and GROM-RD (0.504, 0.510), implying higher sensitivity than other methods.Overall, at each sequencing depth from low to high, LUMPY, CNVnator and GROM-RD had the best performance with respect to the sensitivity of CNV detection.'In Results: Sensitivity and Specificity of CNV prediction-third paragraph: 'The FDR value of iCopyDAV reached a peak value at a 30X depth (0.878), followed by CNVnator (0.767) and RDXplorer (0.731), but these three methods also predicted the most CNVs (Fig 2B).' Should be 'The FDR value of Rsicnv reached a peak value at a 5X depth (0.812), followed by cn.MOPS (0.793) and CNVnator (0.767), but CNVnator also predicted the most CNVs (Fig 2B).'In the Discussion-third paragraph: 'Since TPR values for most methods were below 0.8 and the FDR values for most methods were above 0.3 (Fig 2), we believe that the sensitivity and specificity for CNV detection are not likely to be improved in the future.'Should be 'Since TPR values for most methods were below 0.8 and the FDR values for most methods were above 0.3 (Fig 2), we believe that the sensitivity and specificity for CNV detection are likely to be improved in the future.'

Motivation: This study addresses several important questions related to naturally underrepresented sequences: (i) are there permutations of real genomic DNA sequences in a defined length (k-mer) and a given lineage that do not actually exist or underrepresented? (ii) If there are such sequences, what are their characteristics in terms of k-mer length and base composition? (iii) Are they related to CpG or TpA underrepresentation known for human sequences? We propose that the answers to these questions are of great significance for the study of sequence-associated regulatory mechanisms, such cytosine methylation and chromosomal structures in physiological or pathological conditions such as cancer. Results: We empirically defined sequences that were not included in any well-known public databases as lineage-associated underrepresented permutations (LAUPs). Then, we developed a Jellyfish-based LAUPs analysis application (JBLA) to investigate LAUPs for 24 representative species. The present discoveries include: (i) lengths for the shortest LAUPs, ranging from 10 to 14, which collectively constitute a low proportion of the genome. (ii) Common LAUPs showing higher CG content over the analysed mammalian genome and possessing distinct CG*CG motifs. (iii) Neither CpG-containing LAUPs nor CpG island sequences are randomly structured and distributed over the genomes; some LAUPs and most CpG-containing sequences exhibit an opposite trend within the same k and n variants. In addition, we demonstrate that the JBLA algorithm is more efficient than the original Jellyfish for computing LAUPs. Availability and implementation: We developed a Jellyfish-based LAUP analysis (JBLA) application by integrating Jellyfish (Marçais and Kingsford, 2011), MEME (Bailey, et al., 2009) and the NCBI genome database (Pruitt, et al., 2007) applications, which are listed as Supplementary Material. Supplementary information: Supplementary data are available at Bioinformatics online.

The complexity of biological data is a difficult challenge for data integration. Thus, this study develops a localized web service platform for the specific bioinformatics study needs of Southwest University of China. Our platform includes the following three innovation features, (1) Graphical Survival time analysis kit; (2) Codon Deviation Coefficient (CDC) computing kit and (3) Long non-coding RNA (lncRNA) analysis kit. These feature offer the convenience to the experimentalists for the biological data analysis.

Currently, colorectal cancer (CRC) already becomes one of the most common cancers worldwide. Though the prognosis of CRC patients is dramatically improved due to the new advanced treatments and medical improvements, the 5-year survival rate for the CRC patient is still low. Thus, we hypothesize that CRC may result from the complicated reasons related to both genetic and environmental factors. For this reason, this study collects such big CRC data with information of genetic variations and environmental exposure for the CRC patients and cancer-free controls that are employed to train and test the predictive CRC model. Our results demonstrate that (1) the explored genetic and environmental biomarkers are validated to cause the CRC by the manually reviewed experimental evidences, (2) the model can efficiently predict the risk of CRC after parameter optimization by the big CRC-related data, (3) our innovated generalized kernel recursive maximum correntropy(GKRMC) algorithm has high predictive power. Finally, we discuss why the GKRMC can outperform the classical regression algorithms and the related future study.

Since Agent based model (ABM) can describe the biological system in detail, it is broadly used for multi-scale immune system modeling. However, ABM requires such a high computing cost for the large scale biological modeling that prevents us employing it for the real time system simulation. For this reason, this study develops an orthogonal polynomial based model to approximate solution of ABM, which can obtain low computing cost and high approximating accuracy.

MOTIVATION: p38 mitogen-activated protein kinase activation plays an important role in resistance to chemotherapeutic cytotoxic drugs in treating multiple myeloma (MM). However, how the p38 mitogen-activated protein kinase signaling pathway is involved in drug resistance, in particular the roles that the various p38 isoforms play, remains largely unknown. METHOD: To explore the underlying mechanisms, we developed a novel systems biology approach by integrating liquid chromatography-mass spectrometry and reverse phase protein array data from human MM cell lines with computational pathway models in which the unknown parameters were inferred using a proposed novel algorithm called modularized factor graph. RESULTS: New mechanisms predicted by our models suggest that combined activation of various p38 isoforms may result in drug resistance in MM via regulating the related pathways including extracellular signal-regulated kinase (ERK) pathway and NFкB pathway. ERK pathway regulating cell growth is synergistically regulated by p38δ isoform, whereas nuclear factor kappa B (NFкB) pathway regulating cell apoptosis is synergistically regulated by p38α isoform. This finding that p38δ isoform promotes the phosphorylation of ERK1/2 in MM cells treated with bortezomib was validated by western blotting. Based on the predicted mechanisms, we further screened drug combinations in silico and found that a promising drug combination targeting ERK1/2 and NFκB might reduce the effects of drug resistance in MM cells. This study provides a framework of a systems biology approach to studying drug resistance and drug combination selection. AVAILABILITY AND IMPLEMENTATION: RPPA experimental Data and Matlab source codes of modularized factor graph for parameter estimation are freely available online at http://ctsb.is.wfubmc.edu/publications/modularized-factor-graph.php.

BACKGROUND: The epidermal growth factor receptor (EGFR) signaling pathway and angiogenesis in brain cancer act as an engine for tumor initiation, expansion and response to therapy. Since the existing literature does not have any models that investigate the impact of both angiogenesis and molecular signaling pathways on treatment, we propose a novel multi-scale, agent-based computational model that includes both angiogenesis and EGFR modules to study the response of brain cancer under tyrosine kinase inhibitors (TKIs) treatment. RESULTS: The novel angiogenesis module integrated into the agent-based tumor model is based on a set of reaction-diffusion equations that describe the spatio-temporal evolution of the distributions of micro-environmental factors such as glucose, oxygen, TGFα, VEGF and fibronectin. These molecular species regulate tumor growth during angiogenesis. Each tumor cell is equipped with an EGFR signaling pathway linked to a cell-cycle pathway to determine its phenotype. EGFR TKIs are delivered through the blood vessels of tumor microvasculature and the response to treatment is studied. CONCLUSIONS: Our simulations demonstrated that entire tumor growth profile is a collective behaviour of cells regulated by the EGFR signaling pathway and the cell cycle. We also found that angiogenesis has a dual effect under TKI treatment: on one hand, through neo-vasculature TKIs are delivered to decrease tumor invasion; on the other hand, the neo-vasculature can transport glucose and oxygen to tumor cells to maintain their metabolism, which results in an increase of cell survival rate in the late simulation stages.