Splice-altering variants (SAVs) are the second most prevalent class of pathogenic genetic variants and are strongly associated with the occurrence and development of various diseases. However, current computational tools exhibit limited predictive capability beyond canonical GT-AG splice sites, making accurate assessment of noncanonical SAV pathogenicity a considerable challenge. To address this limitation, we developed MOSAIC (multimodal feature fusion for noncanonical splice-altering variants pathogenicity prediction), a deep learning framework designed for precise assessment of noncanonical SAV pathogenicity. MOSAIC integrates long-range contextual signals derived from a pretrained DNA language model, local sequence features captured from multi-scale convolutional neural networks, and functional annotations. By employing a transformer encoder and a gated fusion module, the model adaptively integrates these multimodal features. Benchmarking across multiple independent datasets demonstrated that MOSAIC consistently outperforms existing state-of-the-art methods, such as CADD and SpliceAI. It remains highly accurate and robust when evaluated on rare variants, gene-independent contexts, and the largest subset where all comparative methods yielded outputs. Furthermore, feature importance analysis revealed that long-range dependencies in DNA sequences and transformer-based integration were critical contributors to model performance. Interpretability analyses indicated that MOSAIC could identify key regulatory sequence motifs associated with transcription factors and RNA-binding proteins, offering mechanistic insight into how noncanonical SAVs disrupt splicing regulation and contribute to pathogenic processes. Overall, MOSAIC offers an accurate and interpretable framework for predicting the pathogenicity of noncanonical SAVs, thereby serving as a dependable computational tool for genetic diagnostics and precision medicine applications. MOSAIC source code and data are available at https://github.com/Lilab-genomics/MOSAIC.

The development of single-cell RNA sequencing (scRNA-seq) technology provides unprecedented opportunities for elucidating cell heterogeneity and gene expression. Identifying and discovering cell types through cell clustering is a crucial step in analyzing scRNA-seq data. However, the high-dimensionality nature and frequent dropout events of the data raise great challenges for cell clustering. Here, we propose a novel contrastive clustering framework called scSCCNIA (Similarity-matrix-based Contrastive Clustering with Neighbor Information Aggregation), for the accurate identification of cell clusters from scRNA-seq data. scSCCNIA adopts a Laplacian filter to conduct neighbor information aggregation, constructs different graph views by using special un-shared parameters Siamese encoders for data augmentation, and learns the latent low-dimensional embedding representations via similarity-matrix-based contrastive learning. Comparative analyses of multiple scRNA-seq datasets from different platforms and with varying cell numbers demonstrate that scSCCNIA outperforms existing methods in terms of cell clustering and marker gene identification. Furthermore, scSCCNIA reveals the heterogeneity and functional specificity of various cell types through Gene Ontology terms and Kyoto Encyclopedia of Genes and Genomes enrichment analyses. Overall, scSCCNIA is an effective algorithm for learning latent features from scRNA-seq data, enhancing cell type identification accuracy and facilitating downstream analyses of scRNA-seq data.

Since genomics was proposed, the exploration of genes has been the focus of research. The emergence of single-cell RNA sequencing (scRNA-seq) technology makes it possible to explore gene expression at the single-cell level. Due to the limitations of sequencing technology, the data contains a lot of noise. At the same time, it also has the characteristics of high-dimensional and sparse. Clustering is a common method of analyzing scRNA-seq data. This paper proposes a novel single-cell clustering method called Robust Manifold Nonnegative Low-Rank Representation with Adaptive Total-Variation Regularization (MLRR-ATV). The Adaptive Total-Variation (ATV) regularization is introduced into Low-Rank Representation (LRR) model to reduce the influence of noise through gradient learning. Then, the linear and nonlinear manifold structures in the data are learned through Euclidean distance and cosine similarity, and more valuable information is retained. Because the model is non-convex, we use the Alternating Direction Method of Multipliers (ADMM) to optimize the model. We tested the performance of the MLRR-ATV model on eight real scRNA-seq datasets and selected nine state-of-the-art methods as comparison methods. The experimental results show that the performance of the MLRR-ATV model is better than the other nine methods.

Single-cell multi-omics sequencing represents an advanced technology capable of simultaneously measuring multiple omics data from the same cell. The joint clustering of single-cell multi-omics sequencing data enables a comprehensive depiction of cell states and uncovers intricate molecular mechanisms, holding immense significance in fields such as oncology, neurology, and developmental biology. However, the disparities in feature spaces across different omics layers and data noise present substantial challenges for achieving accurate clustering. To tackle these challenges, we introduce a novel clustering method for single-cell multi-omics data, termed scMSAC, which is grounded in a denoising subgraph attention autoencoder. The proposed method employs a weighted nearest neighbor graph strategy to ascertain the weights of multi-omics data, subsequently generating a similarity graph that holistically encapsulates intercellular connections through the weighted amalgamation of diverse omics perspectives. The scMSAC model captures the topological features of cells through the subgraph attention autoencoder, constructing relationships among cells. For the omics features extracted by the subgraph attention autoencoder, scMSAC incorporates an SCA (Spatial Channel Attention) mechanism for feature fusion to reduce the differences in feature spaces of different omics and achieve better clustering performance. Comparative experiments with various existing methods demonstrate that scMSAC has excellent clustering performance and performs well in detecting rare cell types and differential expression analysis.

Graph transformer networks have received more attention in hyperspectral image (HSI) classification. However, they overlooked the influence of graph connectivity strength in positional encoding and distribution. In order to address the above deficiencies, we proposed the novel graph transformer with structural embedding and training (GTSET) for HSI classification. Specifically, the structural embedding module firstly aimed at extracting effectively local and non-local feature information via patch-based distance encoding and centrality correlation coefficients based on graph connectivity strength, alleviating spectral variability. Secondly, the structural training module aimed at addressing imbalanced structural position distribution of labeled samples by leveraging the topological graph connectivity to determine their structural position distribution and reweighting the influence of labeled samples on the graph transformer training stage, exploring the guiding role of labeled samples in low spatial resolution of HSI. Next, we further refine training weights based on the spectral feature smoothness of labeled samples. Finally, comprehensive experiments on three real-world HSI datasets demonstrate that the GTSET achieves superior performance in HSI classification with limited labeled samples, compared to other popular classification methods. Implementation of GTSET, along with examples, can be found on the GitHub repository: https://github.com/xuchengchao0/GTSET.

Predicting drug-disease associations is a crucial step in drug repositioning, especially with computational methods that quickly locate potential drug-disease pairs. Heterogenous network is a common tool for introducing multiple type relation information about drugs and diseases. However, the diversity of relations is ignored in most of existing methods, which makes them difficult to explore type semantic information with structure properties. Therefore, we propose a relation-centric GNN framework to encode critical association patterns. Firstly, we utilize a relation-centric graph, line graph, to represent the context of a drug-disease pair identified as the center node. The prediction problem is modeled to learn the embedding vector of the center node. Secondly, a multi-grained line graph neural network (MGLGNN) is designed to excavate fine-grained features that encapsulate local graph structures. We theoretically define a handful of typical nodes that can be regarded as high-order abstractions of relations in each type. Then, MGLGNN distills the local information and passes it to typical nodes from a global perspective. With learned multi-grained features, the center node automatically captures heterogenous relation semantics and structure patterns. Thirdly, a hierarchical contrastive learning (HCL) mechanism is proposed to ensure the quality of multi-grained features in an unsupervised way. Extensive experiments show the great potential of our model in mining drug-disease associations.

Convolutional Neural Networks struggle with long-range dependencies modeling in medical image segmentation, and traditional Transformer models rely on Multi-Layer Perceptron (MLP) for channel information mixing, with performance issues as data dimensions increase. These issues prompt a reassessment of the model's design to enhance segmentation performance and effectively capture long-range dependencies. Consequently, this study presents the Kadformer, a novel network optimized for fine-grained multi-organ segmentation. The Kadformer model adopts an innovative U-shaped network architecture, which enhances the extraction of spatial and channel features in the encoder through the KAN-Enhanced Multi-Dimensional Attention (KMA) mechanism, effectively compensating for information loss during downsampling. We design a Dynamic Path Selection (DPS) strategy to mitigate the feature extraction discrepancies encountered by the linear attention mechanism when processing category-sparse and category-dense images while enhancing feature discrimination through long-range sequential modeling Mamba. Furthermore, we construct the Data Interaction (DAI) module to guide the dual-path encoder's channel and spatial information filtering and effectively integrate the semantically inconsistent features between the KMA and DPS modules. Our approach achieves more than 30% parameter reduction compared to state-of-the-art methods. In addition, the Kadformer network outperforms existing segmentation methods on six public datasets, demonstrating excellent performance. The code has been made available on GitHub: https://github.com/wxc9927/Kadformer.

Single-cell RNA sequencing (scRNA-seq) data typically exhibit high dimensionality, sparsity, and noise, posing significant challenges for downstream tasks such as clustering and biomarker discovery. To address these issues, we propose a Self-Paced Framework Fusing Feature Selection and Clustering with Prior Knowledge (SPF-FS), which tightly integrates feature selection and clustering in an iterative feedback loop. SPF-FS constructs multiple base learners that, together with Gene Ontology information, assess gene importance. We design a joint optimization objective with two goals: enforcing intracluster compactness and maximizing inter-cluster separation. Through a self-paced strategy, the framework incrementally incorporates more challenging sample pairs at each iteration, simultaneously updating feature weights and cluster assignments. This design reduces noisy genes and enhances the selection of biologically relevant features. Comprehensive evaluations on 16 public scRNA-seq datasets show that SPF-FS outperforms nine mainstream methods in terms of the Adjusted Rand Index and normalized mutual information, with average improvements of 0.17 in ARI and 0.18 in NMI. Moreover, SPF-FS exhibits strong stability and biological interpretability in marker-gene selection, and functional pathway enrichment analyses.

Single-cell RNA sequencing (scRNA-seq) deciphers cellular heterogeneity at single-cell resolution, unmasking rare and dynamic cell types. Existing deep learning based methods for single-cell clustering mainly emphasize instance-level representations, while neglecting fundamental clustering principles: maximizing intra-population homogeneity and minimizing interpopulation similarity. Consequently, this oversight constrains clustering performance and compromises the accuracy of cell identification. To address these limitations, we propose sc3M, a novel multi-level graph contrastive learning framework that integrates instance-, cluster- and global-level contrastive learning for scRNA-seq data clustering. sc3M generates three augmented views (original, topological and semantic), utilizes Graph Attention Networks (GAT) and MLP to learn hierarchical cell representations, and enforces multi-level min-max mutual information to optimize cross-view consistency. We evaluated sc3M on nine scRNA-seq datasets. The extensive experiments demonstrated that$\operatorname{sc3M}$achieves superior clustering performance over other state-of-the-art methods. Furthermore, downstream analyses including marker gene identification and cell trajectory inference validate its biological utility. The source code is available at: https://github.com/Biostar1099/sc3M.

Infectious diseases continue to pose a serious threat to public health, underscoring the urgent need for effective computational approaches to screen novel anti-infective agents. Oligopeptides have emerged as promising candidates in an-timicrobial research due to their structural simplicity, high bioavailability, and low susceptibility to resistance. Despite their potential, computational models specifically designed to predict associations between oligopeptides and infectious diseases remain scarce. We propose Prompt-Guided Graph Contrastive Learning for Oligopeptide-Disease Association Prediction (PGCLODA), a framework designed to predict potential associations between oligopeptides and infectious diseases. A tripartite graph is constructed to integrate oligopeptides, microbes, and diseases. To preserve informative structures, we introduce a prompt-based graph augmentation strategy. Our dual encoder, composed of a graph convolutional network and a Transformer, captures both local and global features. The fused representations are used for final classification. Experiments on a benchmark dataset show that our method consistently outperforms existing models and demonstrates strong generalization capability in discovering novel associations.

In multi-organ segmentation tasks, both local details and global contextual information are crucial. Existing main-stream methods based on CNN-Transformer hybrid architectures typically employ simple serial stacking, end-stage concatenation, or pointwise addition for feature fusion, which struggle to handle feature inconsistency and often lead to information conflict and loss. To address the aforementioned challenges, we innovatively propose TBHF-Unet. We design a three-branch hierarchical encoder that dynamically fuses multi-source features in parallel, achieving deep layer-wise integration of multi-source information. The hierarchical structure maintains the independence of each branch while avoiding feature degradation, enabling superior performance without the need for excessively deep networks. Additionally, we design a Local-Global Feature Fusion (LGFF) module to efficiently and accurately integrate local details with global semantics, effectively alleviating feature inconsistency and achieving more comprehensive feature representation. Experiments on five public datasets demonstrate that the proposed method outperforms existing segmentation techniques, showing higher segmentation accuracy and robustness.

Nowadays, technological advancements enable the simultaneous profiling of the epigenome, transcriptome, and proteome at the single-cell level, and an urgent need for tools to integrate such complex tri-modal data is demanded. In this paper, we propose a novel method named scHCHF, a scalable and unified Hierarchical Coarsening Hypergraph Framework. scHCHF first employs a hierarchical coarsening process to groups cells into hypervertices. A unified multiomics hypergraph is then constructed upon these hypervertices, which substantially reduces computational complexity and memory footprint. Subsequently, a hypergraph attention network captures high-order relationships and learns discriminative cell representations, guided by a weakly supervised prototypical contrastive loss. Building upon the pre-trained network, scHCHF further employs a fine-tuning paradigm to achieve accurate cell type annotation using a minimal number of labeled cells. Experiments on real tri-modal datasets demonstrate that scHCHF outperforms other state-of-the-art methods in both cell clustering and cell type annotation tasks.

Single-cell RNA sequencing (scRNA-seq) enables the reconstruction of cell type-specific gene regulatory networks (GRNs), offering detailed insights into gene regulation at high resolution. While graph neural networks have become widely used for GRN inference, their message-passing mechanisms are often limited by issues such as over-smoothing and over-squashing, which hinder the preservation of essential network structure. To address these challenges, we propose a novel graph transformer-based model, AttentionGRN, which leverages soft encoding to enhance model expressiveness and improve the accuracy of GRN inference from scRNA-seq data. Furthermore, the GRN-oriented message aggregation strategies are designed to capture both the directed network structure information and functional information inherent in GRNs. Specifically, we design directed structure encoding to facilitate the learning of directed network topologies and employ functional gene sampling to capture key functional modules and global network structure. Our extensive experiments, conducted on 88 datasets across two distinct tasks, demonstrate that AttentionGRN consistently outperforms existing methods. Furthermore, AttentionGRN has been successfully applied to reconstruct cell type-specific GRNs for human mature hepatocytes, revealing novel hub genes and previously unidentified transcription factor-target gene regulatory associations.

Single-cell RNA sequencing (scRNA-seq) provides transcriptome profiling of individual cells, allowing for in-depth studies of cell heterogeneity at cell resolution. While cell clustering lays the basic foundation of scRNA-seq data analysis, the high-dimensionality and frequent dropout events of the data raise great challenges. Although plenty of dedicated clustering methods have been proposed, they often fail to fully explore the underlying data structure. Here, we introduce scMCGF, a new multi-view clustering algorithm based on graph fusion. It utilizes multi-view data generated from transcriptomic data to learn the consistent and complementary information across different view, ultimately constructing a unified graph matrix for robust cell clustering. Specifically, scMCGF utilizes two-dimensional-reduction methods (principal component analysis and diffusion maps) to capture both linear and non-linear characteristics of the data. Additionally, it calculates a cell-pathway score matrix to incorporate pathway-level information. These three features, along with the pre-processed gene expression data, form the multi-view data. scMCGF iteratively refines the structure of similarity graphs of each view through adaptive learning and learns a unified graph matrix by weighting and fusing the individual similarity graph matrix. The final clustering results are obtained by applying the rank constraint on the Laplacian matrix of the unified graph matrix. Experiments results of 13 real data sets reveal that scMCGF outperforms eight state-of-the-art methods in clustering accuracy and robustness. Furthermore, biological analysis validates that the clustering results of scMCGF provide a reliable foundation for downstream investigations.

Reconstructing high-resolution gene regulatory networks (GRNs) based on single-cell RNA sequencing data provides an opportunity to gain insight into disease pathogenesis. At present, there are a large number of GRN reconstruction methods based on graph neural networks, and they can obtain excellent performance in GRN inference by extracting network structure features. However, most of these methods fail to fully exploit the directional characteristics or even ignore them when extracting network structural features. To this end, a novel framework called GAEDGRN is proposed based on gravity-inspired graph autoencoder (GIGAE) to infer potential causal relationships between genes. Among them, GIGAE can help us capture the complex directed network topology in GRN. Additionally, due to the uneven distribution of the latent vectors generated by the graph autoencoder, a random walk-based method is used to regularize the latent vectors learnt by the encoder. Furthermore, considering that some genes in GRN usually have a significant impact on biological functions, GAEDGRN designs a gene importance score calculation method and pays attention to genes with high importance in the process of GRN reconstruction. Experimental results on seven cell types of three GRN types show that GAEDGRN achieves high accuracy and strong robustness. Moreover, a case study on human embryonic stem cells demonstrates that GAEDGRN can help identify important genes.

Molecular optimization, aiming to identify molecules with improved properties from a huge chemical search space, is a critical step in drug development. This task is challenging due to the need to optimize multiple properties while adhering to stringent drug-like criteria. Recently, numerous effective artificial intelligence methods have been proposed for molecular optimization. However, most of them neglect the constraints in molecular optimization, thereby limiting the development of high-quality molecules that simultaneously satisfy property objectives and constraint compliance. To address this issue, we proposed a deep multi-objective optimization framework, termed CMOMO, for constrained molecular multi-property optimization. The proposed CMOMO divides the optimization process into two stages, which enables it to use a dynamic constraint handling strategy to balance multi-property optimization and constraint satisfaction. Besides, a latent vector fragmentation based evolutionary reproduction strategy is designed to generate promising molecules effectively. Experimental results on two benchmark tasks show that the proposed CMOMO outperforms five state-of-the-art methods to obtain more successfully optimized molecules with multiple desired properties and satisfying drug-like constraints. Moreover, the superiority of CMOMO is verified on two practical tasks, including a potential protein-ligand optimization task of 4LDE protein, which is the structure of $\beta $2-adrenoceptor GPCR receptor, and a potential inhibitor optimization task of glycogen synthase kinase-3$\beta $ target (GSK3$\beta $). Notably, CMOMO demonstrates a two-fold improvement in success rate for the GSK3$\beta $ optimization task, successfully identifying molecules with favorable bioactivity, drug-likeness, synthetic accessibility, and adherence to structural constraints.

Cell type annotation is a critical step in analyzing single-cell RNA sequencing (scRNA-seq) data. A large number of deep learning (DL)-based methods have been proposed to annotate cell types of scRNA-seq data and have achieved impressive results. However, there are several limitations to these methods. First, they do not fully exploit cell-to-cell differential features. Second, they are developed based on shallow features and lack of flexibility in integrating high-order features in the data. Finally, the low-dimensional gene features may lead to overfitting in neural networks. To overcome those limitations, we propose a novel DL-based model, cell type annotation of single-cell RNA-seq data using residual graph convolutional neural network with contrastive learning (scRGCL), based on residual graph convolutional neural network and contrastive learning for cell type annotation of single-cell RNA-seq data. scRGCL mainly consists of a residual graph convolutional neural network, contrastive learning, and weight freezing. A residual graph convolutional neural network is utilized to extract complex high-order features from data. Contrastive learning can help the model learn meaningful cell-to-cell differential features. Weight freezing can avoid overfitting and help the model discover the impact of specific gene expression on cell type annotation. To verify the effectiveness of scRGCL, we compared its performance with six methods (three shallow learning algorithms and three state-of-the-art DL-based methods) on eight single-cell benchmark datasets from two species (seven in human and one in mouse). Experimental results not only show that scRGCL outperforms competing methods but also demonstrate the generalizability of scRGCL for cell type annotation. scRGCL is available at https://github.com/nathanyl/scRGCL.

BACKGROUND: The progression of cancer is driven by the accumulation of mutations in driver genes. Many researches promote to identify cancer driver genes. However, most of them ignore the high-order features in the network. RESULT: In this study, we propose a novel method MLGCN-Driver based on multi-layer graph convolutional neural networks (GCN) to boost driver gene identification. MLGCN-Driver employs multi-layer GCN with initial residual connections and identity mappings to learn biological multi-omics features within biological networks. In addition, node2vec algorithm is used to extract the topological structure features of the biological network, and then the features are fed into another multi-layer GCN for feature learning. Meanwhile, the initial residual connections and identity mappings mitigate the over-smooth of features. Finally, the probability of each gene being a driver gene is calculated based on low-dimensional biological features and topological features. CONCLUSION: We applied the MLGCN-Driver on pan-cancer dataset and cancer type-specific datasets. Experimental results demonstrate the excellent performance of MLGCN-Driver in terms of the area under the ROC curve (AUC) and the area under the precision-recall curve (AUPRC) when compared with state-of-the-art approaches.

As a fundamental characteristic of multicellular organisms, cell-cell communication is achieved through ligand-receptor (L-R) interactions, enabling the exchange of information and revealing the diversity of biological processes and cellular functions. To gain a comprehensive understanding of these complex interaction mechanisms, we constructed a manually curated L-R interaction database and developed a semi-supervised graph embedding model called scSDNE for inferring cell-cell interactions mediated by L-R interactions. scSDNE model utilizes the power of deep learning to map genes from interacting cells into a shared latent space, allowing for a nuanced representation of their relationships. Leveraging the prior information provided by database, scSDNE can infer significant L-R pairs involved in intercellular communication. Experiments on real single-cell RNA sequencing (scRNA-seq) datasets demonstrate that our method detects interactions with a high degree of reliability compared with other methods. More importantly, the model integrates gene regulation information within cells to enhance the accuracy and biological interpretability of the inferences. Our method provides a more comprehensive view of cell-cell interactions, offering new insights into complex intercellular communication.

Cancer drug combination therapies offer a promising strategy to overcome resistance and improve treatment efficacy, but identifying synergistic drug pairs is challenging due to complex biological interactions and tumor heterogeneity. Current machine learning algorithms for drug synergy prediction primarily rely on large-scale, multimodal datasets, yet suffer from critical limitations including poor interpretability, difficulty distinguishing causative biological relationships from correlations, and inadequate modeling of cancer-specific molecular interactions. To address these challenges, we propose CASynergy (Causal Attention and Cross-attention Synergy), a novel deep learning model for predicting cancer drug synergy that addresses limitations of prior approaches in accuracy and interpretability. CASynergy introduces a causal attention mechanism to distinguish true causal genomic features from spurious correlations, cell line-specific gene network construction to capture the unique molecular context of each cancer cell line, and a cross-attention module to integrate drug molecular features with cell line gene expression profiles. These improvements allow CASynergy to clearly identify significant drug-gene interactions and provides interpretable insights into why a combination is predicted to be synergistic. Experiments on two benchmark datasets (DrugCombDB and Oncology-Screen) suggests that CASynergy outperformed five state-of-the-art models. CASynergy offers a better and more reliable way to predict effective drug combinations. It works well across different cancer types and is easier to understand, which is important for personalized cancer treatment and finding new drugs.

Classification methods based on subgraph neural networks (SNNs) are rarely explored, and its advantage is that it can alleviate the neighbor explosion problem. After applying SNNs to hyperspectral image (HSI) classification, the imbalanced topology structure in the internal subgraph leads to poor classification performance due to the intraclass and interclass spectral feature variation. Based on this, we proposed a novel subgraph-dependent neural network (SGDNet) for HSI classification. Specifically, we firstly segmented the large graph to a series of subgraphs, and proposed a subgraph-dependent convolution method for imbalanced subgraph structure to achieve effective feature smoothing within each subgraph. It mainly utilized the degree feature embedding with the residual to improve the feature diversity as well as the intraclass distance of clustering measurement to determine the optimal subgraph convolution layers in a feedback way. Secondly, we developed the strategy of the long-range dependency to address the inevitably local structure dependence among nodes within subgraphs. It utilized an anchor point based position coding method to capture the relative positions of unlabeled nodes in relation to all labeled nodes within the graph and further constructed the structural loss function for effective training, achieving subgraph optimization. Comprehensive experiments demonstrated the superior performance of the SGDNet model on three publicly available HSI datasets, compared to other popular methods. The source code will be available at https://github.com/lichao226211/SGDNet.

Intracranial aneurysm (IA) is a vascular disease of the brain arteries caused by pathological vascular dilation, which can result in subarachnoid hemorrhage if ruptured. Automatically classification and segmentation of intracranial aneurysms are essential for their diagnosis and treatment. However, the majority of current research is focused on two-dimensional images, ignoring the 3D spatial information that is also critical. In this work, we propose a novel dual-branch fusion network called the Point Cloud and Multi-View Medical Neural Network (PMMNet) for IA classification and segmentation. Specifically, one branch based on 3D point clouds serves the purpose of extracting spatial features, whereas the other branch based on multi-view images acquires 2D pixel features. Ultimately, the two types of features are fused for IA classification and segmentation. To extract both local and global features from 3D point clouds, Multilayer Perceptron (MLP) and the attention mechanism are used in parallel. In addition, a SPSA module is proposed for multi-view image feature learning, which extracts more exquisite channel and spatial multi-scale features from 2D images. Experiments conducted on the IntrA dataset outperform other state-of-the-art methods, demonstrating that the proposed PMMNet exhibits strong superiority on the medical 3D dataset. We also obtain competitive results on public datasets, including ModelNet40, ModelNet10, and ShapeNetPart, which further validate the robustness and generality of the PMMNet.

Drug repositioning (DR) has emerged as an effective method of identifying new indications for existing drugs. Many DR methods have demonstrated superior performance. However, most of them utilize a limited number of biological entities, ignoring the critical role of other entities in addressing data sparsity as well as improving model generalization capabilities. In addition, fully capturing high-order information of biological data still needs to be fully explored. To address above issues, a model based on transformer and enhanced multi-view contrastive learning (TEMCL) is proposed for predicting drug-disease associations (DDAs). Firstly, transformer is employed to obtain high-order features of nodes from similarity information. Secondly, based on similarity matrices and association matrices of nodes, two different types of views are constructed, i.e., homogeneous hypergraphs and heterogeneous association graphs. Among them, to alleviate sparsity problem existing in heterogeneous graphs, protein nodes as well as meta-path enhancement strategy are introduced. Thirdly, hypergraph convolutional network and heterogeneous graph transformer are used to extract node features on above two types of views, respectively. Contrastive learning is applied to obtain more representative features. Finally, multilayer perceptron (MLP) is used for predicting DDAs. Experiments show that TEMCL outperforms existing methods on DR task, exhibiting superior performance. In addition, case studies further demonstrate the effectiveness of this model. TEMCL provides new insights for identifying novel DDAs.

Synergistic drug combinations have shown promising results in treating cancer cell lines by enhancing therapeutic efficacy and minimizing adverse reactions. The effects of a drug vary across cell lines, and cell lines respond differently to various drugs during treatment. Recently, many AI-based techniques have been developed for predicting synergistic drug combinations. However, existing computational models have not addressed this phenomenon, neglecting the refinement of features for the same drug and cell line in different scenarios. In this work, we propose a feature refinement deep learning framework, termed FRSynergy, to identify synergistic drug combinations. It can guide the refinement of drug and cell line features in different scenarios by capturing relationships among diverse drug-drug-cell line triplet features and learning feature contextual information. The heterogeneous graph attention network is employed to acquire topological information-based original features for drugs and cell lines from sampled sub-graphs. Then, the feature refinement network is designed by combining attention mechanism and context information, which can learn context-aware feature representations for each drug and cell line feature in diverse drug-drug-cell line triplet contexts. Extensive experiments affirm the strong performance of FRSynergy in predicting synergistic drug combinations and, more importantly, demonstrate the effectiveness of feature refinement network in synergistic drug combination prediction.

Segmentation of thoracic and abdominal organs is crucial for accurate disease diagnosis, surgical planning, and long-term health management of patients. Deep learning often depends on large quantities of high-quality training data to achieve superior results. However, the inherent complexity and sparsity of medical images require advanced models with greater learning capabilities. Additionally, significant morphological differences between organs often result in inaccurate and false-positive segmentation. To alleviate these issues, we propose a Dual Attention (DA) and Multi-Resolution Inputs (MI) based network (DAMNet) for the segmentation of thoracic and abdominal organs. Specially, DAMNet utilizes a single encoder and dual decoders. The encoder integrates MI with the Transformer to achieve multi-level feature fusion and capture global image relationships, improving the model’s capability to process intricate image data. In the decoders, Residual U-blocks (RSU) and the DA module consisting of Spatial Multi-Scale Cross-Axis Attention (SMCA) and Convolutional Self-Attention (CSA) are used as two decoder branches, respectively. The design of decoders allows the model to extract detailed information from different encoding layers from two perspectives, thereby reducing inaccuracies in segmentation. We perform thorough experiments and evaluations using three publicly available datasets: Synapse, SegTHOR, and THoracic. The experimental results indicate that our proposed DAMNet model demonstrates exceptional proficiency in segmenting thoracic and abdominal organs.

Recent developments in spatial transcriptomics technologies have made it possible to obtain gene expression profiles while maintaining spatial context. Precisely identifying spatial domains is essential for downstream analysis, requiring the effective integration of gene expression profiles with spatial information. To overcome the challenge of low accuracy in spatial domain identification, this paper proposed a deep learning model called LGAGCN based on local and global information. It used graph convolutional network to learn the features of local and global views and employed an attention mechanism to integrate embeddings from different views. Moreover, experiments were conducted on the human dorsolateral prefrontal cortex (DLPFC) dataset and the human breast cancer (HBC) dataset to evaluate the effectiveness of the model. The experimental results showed that LGAGCN outperformed state-of-the-art methods in spatial clustering task.

In diagnosing Alzheimer’s Disease (AD), structural Magnetic Resonance Imaging (sMRI) is crucial, providing axial, coronal, and sagittal views for a comprehensive analysis of brain structures and pathological changes. Region of Interest (ROI) techniques in neuroimaging further allow detailed examination of structural and functional changes linked to AD. Additionally, certain SNPs have been identified as significant genetic markers for AD risk. Given the capabilities of Convolutional Neural Networks (CNNs) and Transformer models in handling multimodal data, this study introduces a novel hybrid network, DAResNet-ViT, for AD diagnosis. The model utilizes 24 key sMRI slices from each view per participant, combined with ROI features and genetic data, for classifying subjects. Data from two Alzheimer’s Disease Neuroimaging Initiative (ADNI) datasets were preprocessed and used to train and validate DAResNet-ViT. The model was evaluated through experiments focusing on AD diagnosis and the prediction of Mild Cognitive Impairment (MCI) conversion, with its effectiveness demonstrated in ablation studies and comparative experiments.

Magnetic resonance imaging (MRI) technology shows significant potential in predicting early pathological changes associated with Alzheimer's disease (AD). However, the complexity of MRI and their multidimensional characteristics pose challenges in model classification tasks, and single-scale features often fail to capture subtle changes in lesion areas effectively. To improve the accuracy of AD predictions, we propose a network model based on multi-scale features and a dual self-attention mechanism (MSDA). This model integrates depthwise separable convolutions with an improved self-attention mechanism, thereby enhancing the classification ability for AD. First, MRI undergo head motion correction, image alignment, and skull stripping to enhance the model's capability to extract features related to AD lesions. Second, we designed a multi-scale convolutional network structure that utilizes depthwise separable convolution kernels of varying sizes, allowing the network to effectively capture multi-scale feature information from MRI and accurately identify lesion areas. Finally, we introduced a dual self-attention module, which includes channel self-attention and spatial self-attention, further augmenting the model's ability to extract lesion features by learning the differences in features across different categories of MRI in both channel and spatial dimensions. Experimental results indicate that the MSDA network model demonstrates exceptional performance in classifying normal controls (NC) and AD within the ADNI dataset, outperforming existing models in classification accuracy, performance, and generalization capability. The accuracy, sensitivity, and specificity of the model reached 97.8%, 96.3%, and 99.4%, respectively.

Alzheimer's disease (AD) is a neurodegenerative disorder that progresses in a slow and irreversible manner. Although many computer-aided methods have been used to diagnose AD, the issue of underutilization of detailed information and features persists. In this study, we propose a new AD diagnostic network (FPRes-Net) that can fully learn the rich information of 3D MRI slices by extracting multi-scale features and feature fusion. Firstly, in order to fully extract multi-scale information, a network structure combining ResNet-50 with feature pyramids was designed. Next, a feature fusion method was designed to reduce noise and increase the importance of important features. Finally, a visually interpretable method called Gradient-weighted Class Activation Mapping (Grad-CAM) was introduced to visualize important feature regions in AD diagnosis. Experimental analysis was conducted on the publicly accessible ADNI-1 dataset, and our proposed FPRes-Net model performed better than other advanced research methods, with an accuracy rate of 99.5%. Our proposed model can be effectively used for clinical diagnosis of AD.

The inference of gene regulatory networks (GRNs) from gene expression profiles has been a key issue in systems biology, prompting many researchers to develop diverse computational methods. However, most of these methods do not reconstruct directed GRNs with regulatory types because of the lack of benchmark datasets or defects in the computational methods. Here, we collect benchmark datasets and propose a deep learning-based model, DeepFGRN, for reconstructing fine gene regulatory networks (FGRNs) with both regulation types and directions. In addition, the GRNs of real species are always large graphs with direction and high sparsity, which impede the advancement of GRN inference. Therefore, DeepFGRN builds a node bidirectional representation module to capture the directed graph embedding representation of the GRN. Specifically, the source and target generators are designed to learn the low-dimensional dense embedding of the source and target neighbors of a gene, respectively. An adversarial learning strategy is applied to iteratively learn the real neighbors of each gene. In addition, because the expression profiles of genes with regulatory associations are correlative, a correlation analysis module is designed. Specifically, this module not only fully extracts gene expression features, but also captures the correlation between regulators and target genes. Experimental results show that DeepFGRN has a competitive capability for both GRN and FGRN inference. Potential biomarkers and therapeutic drugs for breast cancer, liver cancer, lung cancer and coronavirus disease 2019 are identified based on the candidate FGRNs, providing a possible opportunity to advance our knowledge of disease treatments.

Prediction of drug-target interactions (DTIs) is essential in medicine field, since it benefits the identification of molecular structures potentially interacting with drugs and facilitates the discovery and reposition of drugs. Recently, much attention has been attracted to network representation learning to learn rich information from heterogeneous data. Although network representation learning algorithms have achieved success in predicting DTI, several manually designed meta-graphs limit the capability of extracting complex semantic information. To address the problem, we introduce an adaptive meta-graph-based method, termed AMGDTI, for DTI prediction. In the proposed AMGDTI, the semantic information is automatically aggregated from a heterogeneous network by training an adaptive meta-graph, thereby achieving efficient information integration without requiring domain knowledge. The effectiveness of the proposed AMGDTI is verified on two benchmark datasets. Experimental results demonstrate that the AMGDTI method overall outperforms eight state-of-the-art methods in predicting DTI and achieves the accurate identification of novel DTIs. It is also verified that the adaptive meta-graph exhibits flexibility and effectively captures complex fine-grained semantic information, enabling the learning of intricate heterogeneous network topology and the inference of potential drug-target relationship.

Cluster assignment is vital to analyzing single-cell RNA sequencing (scRNA-seq) data to understand high-level biological processes. Deep learning-based clustering methods have recently been widely used in scRNA-seq data analysis. However, existing deep models often overlook the interconnections and interactions among network layers, leading to the loss of structural information within the network layers. Herein, we develop a new self-supervised clustering method based on an adaptive multi-scale autoencoder, called scAMAC. The self-supervised clustering network utilizes the Multi-Scale Attention mechanism to fuse the feature information from the encoder, hidden and decoder layers of the multi-scale autoencoder, which enables the exploration of cellular correlations within the same scale and captures deep features across different scales. The self-supervised clustering network calculates the membership matrix using the fused latent features and optimizes the clustering network based on the membership matrix. scAMAC employs an adaptive feedback mechanism to supervise the parameter updates of the multi-scale autoencoder, obtaining a more effective representation of cell features. scAMAC not only enables cell clustering but also performs data reconstruction through the decoding layer. Through extensive experiments, we demonstrate that scAMAC is superior to several advanced clustering and imputation methods in both data clustering and reconstruction. In addition, scAMAC is beneficial for downstream analysis, such as cell trajectory inference. Our scAMAC model codes are freely available at https://github.com/yancy2024/scAMAC.

Inferring gene regulatory network (GRN) is one of the important challenges in systems biology, and many outstanding computational methods have been proposed; however there remains some challenges especially in real datasets. In this study, we propose Directed Graph Convolutional neural network-based method for GRN inference (DGCGRN). To better understand and process the directed graph structure data of GRN, a directed graph convolutional neural network is conducted which retains the structural information of the directed graph while also making full use of neighbor node features. The local augmentation strategy is adopted in graph neural network to solve the problem of poor prediction accuracy caused by a large number of low-degree nodes in GRN. In addition, for real data such as E.coli, sequence features are obtained by extracting hidden features using Bi-GRU and calculating the statistical physicochemical characteristics of gene sequence. At the training stage, a dynamic update strategy is used to convert the obtained edge prediction scores into edge weights to guide the subsequent training process of the model. The results on synthetic benchmark datasets and real datasets show that the prediction performance of DGCGRN is significantly better than existing models. Furthermore, the case studies on bladder uroepithelial carcinoma and lung cancer cells also illustrate the performance of the proposed model.

The gene regulatory network (GRN) plays a vital role in understanding the structure and dynamics of cellular systems, revealing complex regulatory relationships, and exploring disease mechanisms. Recently, deep learning (DL)-based methods have been proposed to infer GRNs from single-cell transcriptomic data and achieved impressive performance. However, these methods do not fully utilize graph topological information and high-order neighbor information from multiple receptive fields. To overcome those limitations, we propose a novel model based on multiview graph attention network, namely, scMGATGRN, to infer GRNs. scMGATGRN mainly consists of GAT, multiview, and view-level attention mechanism. GAT can extract essential features of the gene regulatory network. The multiview model can simultaneously utilize local feature information and high-order neighbor feature information of nodes in the gene regulatory network. The view-level attention mechanism dynamically adjusts the relative importance of node embedding representations and efficiently aggregates node embedding representations from two views. To verify the effectiveness of scMGATGRN, we compared its performance with 10 methods (five shallow learning algorithms and five state-of-the-art DL-based methods) on seven benchmark single-cell RNA sequencing (scRNA-seq) datasets from five cell lines (two in human and three in mouse) with four different kinds of ground-truth networks. The experimental results not only show that scMGATGRN outperforms competing methods but also demonstrate the potential of this model in inferring GRNs. The code and data of scMGATGRN are made freely available on GitHub (https://github.com/nathanyl/scMGATGRN).

MOTIVATION: In the drug development process, a significant portion of the budget and research time are dedicated to the lead compound optimization procedure to identify potential drugs. This procedure focuses on enhancing the pharmacological and bioactive properties of compounds by optimizing their local substructures. However, due to the vast and discrete chemical structure space and the unpredictable element combinations within this space, the optimization process is inherently complex. Various structure enumeration-based combinatorial optimization methods have shown certain advantages. However, they still have limitations. Those methods fail to consider the differences between molecules and struggle to explore the unknown outer search space. RESULTS: In this study, we propose an adaptive space search-based molecular evolution optimization algorithm (ASSMOEA). It consists of three key modules: construction of molecule-specific search space, molecular evolutionary optimization, and adaptive expansion of molecule-specific search space. Specifically, we design a fragment similarity tree in a molecule-specific search space and apply a dynamic mutation strategy in this space to guide molecular optimization. Then, we utilize an encoder-encoder structure to adaptively expand the space. Those three modules are circled iteratively to optimize molecules. Our experiments demonstrate that ASSMOEA outperforms existing methods in terms of molecular optimization. It not only enhances the efficiency of the molecular optimization process but also exhibits a robust ability to search for correct solutions. AVAILABILITY AND IMPLEMENTATION: The code is freely available on the web at https://github.com/bbbbb-b/MEOAFST.

Abstract MicroRNAs (miRNAs) are closely related to numerous complex human diseases, therefore, exploring miRNA-disease associations (MDAs) can help people gain a better understanding of complex disease mechanism. An increasing number of computational methods have been developed to predict MDAs. However, the sparsity of the MDAs may hinder the performance of many methods. In addition, many methods fail to capture the nonlinear relationships of miRNA-disease network and inadequately leverage the features of network and neighbor nodes. In this study, we propose a deep matrix factorization model with variational autoencoder (DMFVAE) to predict potential MDAs. DMFVAE first decomposes the original association matrix and the enhanced association matrix, in which the enhanced association matrix is enhanced by self-adjusting the nearest neighbor method, to obtain sparse vectors and dense vectors, respectively. Then, the variational encoder is employed to obtain the nonlinear latent vectors of miRNA and disease for the sparse vectors, and meanwhile, node2vec is used to obtain the network structure embedding vectors of miRNA and disease for the dense vectors. Finally, sample features are acquired by combining the latent vectors and network structure embedding vectors, and the final prediction is implemented by convolutional neural network with channel attention. To evaluate the performance of DMFVAE, we conduct five-fold cross validation on the HMDD v2.0 and HMDD v3.2 datasets and the results show that DMFVAE performs well. Furthermore, case studies on lung neoplasms, colon neoplasms, and esophageal neoplasms confirm the ability of DMFVAE in identifying potential miRNAs for human diseases.

Circular RNAs (circRNAs) play vital roles in transcription and translation. Identification of circRNA-RBP (RNA-binding protein) interaction sites has become a fundamental step in molecular and cell biology. Deep learning (DL)-based methods have been proposed to predict circRNA-RBP interaction sites and achieved impressive identification performance. However, those methods cannot effectively capture long-distance dependencies, and cannot effectively utilize the interaction information of multiple features. To overcome those limitations, we propose a DL-based model iCRBP-LKHA using deep hybrid networks for identifying circRNA-RBP interaction sites. iCRBP-LKHA adopts five encoding schemes. Meanwhile, the neural network architecture, which consists of large kernel convolutional neural network (LKCNN), convolutional block attention module with one-dimensional convolution (CBAM-1D) and bidirectional gating recurrent unit (BiGRU), can explore local information, global context information and multiple features interaction information automatically. To verify the effectiveness of iCRBP-LKHA, we compared its performance with shallow learning algorithms on 37 circRNAs datasets and 37 circRNAs stringent datasets. And we compared its performance with state-of-the-art DL-based methods on 37 circRNAs datasets, 37 circRNAs stringent datasets and 31 linear RNAs datasets. The experimental results not only show that iCRBP-LKHA outperforms other competing methods, but also demonstrate the potential of this model in identifying other RNA-RBP interaction sites.

MicroRNAs (miRNA) are endogenous non-coding RNAs, typically around 23 nucleotides in length. Many miRNAs have been founded to play crucial roles in gene regulation though post-transcriptional repression in animals. Existing studies suggest that the dysregulation of miRNA is closely associated with many human diseases. Discovering novel associations between miRNAs and diseases is essential for advancing our understanding of disease pathogenesis at molecular level. However, experimental validation is time-consuming and expensive. To address this challenge, numerous computational methods have been proposed for predicting miRNA-disease associations. Unfortunately, most existing methods face difficulties when applied to large-scale miRNA-disease complex networks. In this paper, we present a novel subgraph learning method named SGLMDA for predicting miRNA-disease associations. For miRNA-disease pairs, SGLMDA samples K-hop subgraphs from the global heterogeneous miRNA-disease graph. It then introduces a novel subgraph representation algorithm based on Graph Neural Network (GNN) for feature extraction and prediction. Extensive experiments conducted on benchmark datasets demonstrate that SGLMDA can effectively and robustly predict potential miRNA-disease associations. Compared to other state-of-the-art methods, SGLMDA achieves superior prediction performance in terms of Area Under the Curve (AUC) and Average Precision (AP) values during 5-fold Cross-Validation (5CV) on benchmark datasets such as HMDD v2.0 and HMDD v3.2. Additionally, case studies on Colon Neoplasms and Triple-Negative Breast Cancer (TNBC) further underscore the predictive power of SGLMDA.

Recent advancements in spatially transcriptomics (ST) technologies have enabled the comprehensive measurement of gene expression profiles while preserving the spatial information of cells. Combining gene expression profiles and spatial information has been the most commonly used method to identify spatial functional domains and genes. However, most existing spatial domain decipherer methods are more focused on spatially neighboring structures and fail to take into account balancing the self-characteristics and the spatial structure dependency of spots. Therefore, we propose a novel model called SpaGCAC, which recognizes spatial domains with the help of an adaptive feature-spatial balanced graph convolutional network named AFSBGCN. The AFSBGCN can dynamically learn the relationship between spatial local topology structures and the self-characteristics of spots by adaptively increasing or declining the weight on the self-characteristics during message aggregation. Moreover, to better capture the local structures of spots, SpaGCAC exploits a local topology structure contrastive learning strategy. Meanwhile, SpaGCAC utilizes a probability distribution contrastive learning strategy to increase the similarity of probability distributions for points belonging to the same category. We validate the performance of SpaGCAC for spatial domain identification on four spatial transcriptomic datasets. In comparison with seven spatial domain recognition methods, SpaGCAC achieved the highest NMI median of 0.683 and the second highest ARI median of 0.559 on the multi-slice DLPFC dataset. SpaGCAC achieved the best results on all three other single-slice datasets. The above-mentioned results show that SpaGCAC outperforms most existing methods, providing enhanced insights into tissue heterogeneity.

Rapid advances in single-cell RNA sequencing (scRNA-seq) have made it possible to characterize cell states at a high resolution view for large scale library. scRNA-seq data contains a great deal of biological information, which can be mainly used to discover cell subtypes and track cell development. However, traditional methods face many challenges in addressing scRNA-seq data with high dimensions and high sparsity. For better analysis of scRNA-seq data, we propose a new framework called MSVGAE based on variational graph auto-encoder and graph attention networks. Specifically, we introduce multiple encoders to learn features at different scales and control for uninformative features. Moreover, different noises are added to encoders to promote the propagation of graph structural information and distribution uncertainty. Therefore, some complex posterior distributions can be captured by our model. MSVGAE maps scRNA-seq data with high dimensions and high noise into the low-dimensional latent space, which is beneficial for downstream tasks. In particular, MSVGAE can handle extremely sparse data. Before the experiment, we create 24 simulated datasets to simulate various biological scenarios and collect 8 real-world datasets. The experimental results of clustering, visualization and marker genes analysis indicate that MSVGAE model has excellent accuracy and robustness in analyzing scRNA-seq data.

The advent of single-cell RNA sequencing (scRNA-seq) technology offers the opportunity to conduct biological research at the cellular level. Single-cell type identification based on unsupervised clustering is one of the fundamental tasks of scRNA-seq data analysis. Although many single-cell clustering methods have been developed recently, few can fully exploit the deep potential relationships between cells, resulting in suboptimal clustering. In this paper, we propose scGAMF, a graph autoencoder-based multi-level kernel subspace fusion framework for scRNA-seq data analysis. Based on multiple top feature sets, scGAMF unifies deep feature embedding and kernel space analysis into a single framework to learn an accurate clustering affinity matrix. First, we construct multiple top feature sets to avoid the high variability caused by single feature set learning. Second, scGAMF uses a graph autoencoder (GAEs) to extract deep information embedded in the data, and learn embeddings including gene expression patterns and cell-cell relationships. Third, to fully explore the deep potential relationships between cells, we design a multi-level kernel space fusion strategy. This strategy uses a kernel expression model with adaptive similarity preservation to learn a self-expression matrix shared by all embedding spaces of a given feature set, and a consensus affinity matrix across multiple top feature sets. Finally, the consensus affinity matrix is used for spectral clustering, visualization, and identification of gene markers. Extensive validation on real datasets shows that scGAMF achieves higher clustering accuracy than many popular single-cell analysis methods.

Advances in high-throughput single-cell RNA sequencing (scRNA-seq) technology have provided more comprehensive biological information on cell expression. Clustering analysis is a critical step in scRNA-seq research and provides clear knowledge of the cell identity. Unfortunately, the characteristics of scRNA-seq data and the limitations of existing technologies make clustering encounter a considerable challenge. Meanwhile, some existing methods treat different features equally and ignore differences in feature contributions, which leads to a loss of information. To overcome limitations, we introduce a weighted distance constraint into the construction of the similarity graph and combine the similarity constraint. We propose the Joint Automatic Weighting Similarity Graph and Low-rank Representation (JAGLRR) clustering method. Evaluating the contributions of each feature and assigning various weight values can increase the significance of valuable features while decreasing the interference of redundant features. The similarity constraint allows the model to generate a more symmetric affinity matrix. Benefitting from that affinity matrix, JAGLRR recovers the original linear relationship of the data more accurately and obtains more discriminative information. The results on simulated datasets and 8 real datasets show that JAGLRR outperforms 11 existing comparison methods in clustering experiments, with higher clustering accuracy and stability.

Single-cell RNA sequencing (scRNA-seq) is a potent advancement for analyzing gene expression at the individual cell level, allowing for the identification of cellular heterogeneity and subpopulations. However, it suffers from technical limitations that result in sparse and heterogeneous data. Here, we propose scVSC, an unsupervised clustering algorithm built on deep representation neural networks. The method incorporates the variational inference into the subspace model, which imposes regularization constraints on the latent space and further prevents overfitting. In a series of experiments across multiple datasets, scVSC outperforms existing state-of-the-art unsupervised and semi-supervised clustering tools regarding clustering accuracy and running efficiency. Moreover, the study indicates that scVSC could visually reveal the state of trajectory differentiation, accurately identify differentially expressed genes, and further discover biologically critical pathways.

This study focuses on building an intelligent decision-making attention mechanism in which the channel relationship and conduct feature maps among specific deep Dense ConvNet blocks are connected to each other. Thus, develop a novel freezing network with a pyramid spatial channel attention mechanism (FPSC-Net) in deep modeling. This model studies how specific design choices in the large-scale data-driven optimization and creation process affect the balance between the accuracy and effectiveness of the designed deep intelligent model. To this end, this study presents a novel architecture unit, which is termed as the "Activate-and-Freeze" block on popular and highly competitive datasets. In order to extract informative features by fusing spatial and channel-wise information together within local receptive fields and boost the representation power, this study constructs a Dense-attention module (pyramid spatial channel (PSC) attention) to perform feature recalibration, and through the PSC attention to model the interdependence among convolution feature channels. We join the PSC attention module in the activating and back-freezing strategy to search for one of the most important parts of the network for extraction and optimization. Experiments on various large-scale datasets demonstrate that the proposed method can achieve substantially better performance for improving the ConvNets representation power than the other state-of-the-art deep models.

Graph convolutional networks (GCNs)-based methods for hyperspectral image (HSI) classification have received more attention due to its flexibility in information aggregation. However, most existing GCN-based methods in HSI community rely on capturing fixed K-hops neighbors for feature information aggregation, which ignores the inherent imbalance in class distributions and fails to achieve optimal feature smoothing through graph convolution operator. It is unreasonable to apply fixed K-hops strategy for feature smoothing in imbalanced classes, as class regions with rich contextual information and those with poor contextual information require to capture different hops neighbors to achieve the optimal feature smoothing. To address this issue, this article proposes a novel approach called class-imbalanced graph convolution smoothing (CIGCS) for HSI classification, which achieves adaptive feature smoothing for imbalanced class regions. Firstly, we construct a semantic block-diagonal graph structure that describes imbalanced semantic class regions by considering label connectivity and spectral Laplacian regularizer. Secondly, we develop the class-imbalanced graph convolution smoothing technique to adaptively aggregate neighbor information for imbalanced class regions based on the decreasing Euclidean distance of samples within each bock-diagonal structure from the perspective of over-smoothing. The choice of adaptive neighbors can be guaranteed by a theoretical upper bound. Finally, the obtained optimal smoothed features are fed into the logistic regression to achieve good classification results. The proposed CIGCS method is evaluated on three real HSI data sets to demonstrate its superiority compared to some popular GCN-based methods.

Researchers have been paying more attention to hyperspectral image (HSI) classification based on semi-supervised superpixel-level graph convolutional networks (SGCNs) due to their aggregation ability of rich contextual information. Although these SGCNs achieve good classification performance, the influence of the positional distributions among labeled superpixels has been overlooked. The locations of labeled superpixels, such as located at class boundaries or centers, exert a substantial influence on the final performance. To address this issue, this article proposed a novel graph neural network (GCN) method with the guidance of positional distributions of labeled superpixels, abbreviated as LPDGCN. Specifically, we first propose to utilize the sparse, low-rank as well as feature smoothness restrictions to optimize the initial superpixel graph structure because the connectivity relationships of labeled superpixels located at class boundaries or centers are easily influenced by spectral variation. Second, in order to effectively determine the positional distributions of labeled superpixels and make full use of the position relationships, we propose to utilize the information conflict from the above topology connectivity to determine the positional distributions of labeled superpixels and develop the reweighted strategy to weaken the influence of labeled superpixels located at class boundaries and strengthen the influence of that located at class centers. Finally, we evaluate the LPDGCN method on four public HSI datasets, demonstrating its superiority over other advanced classification methods in terms of three metrics, i.e., overall accuracy (OA), average accuracy (AA), and kappa coefficient (KC).

Meter pointers exhibit stable and anti-interference capabilities, rendering them extensively utilized in industrial environments. However, automated reading poses a significant challenge due to the fact that current segmentation methods struggle to isolate the fine-grained pointers and scales for accurate reading calculations. This challenge can be alleviated by enhancing the feature extraction capability of the segmentation network. As is well-known that Attention plays an essential role in human vision by selectively focusing on convex parts, and attention-based methods have been applied to various computer vision tasks. Therefore, we propose a new image segmentation network called multipath attention network (MPANet) for pointer meter recognition in the complex industrial environments. The designed network employs an attention gate mechanism to proficiently capture local features stemming from various pathways during skip-connection and upsample processes. In addition, our network incorporates deep supervision by merging the outputs of the final three layers to extract abundant low-dimensional information. To further improve the performance of encoders and decoders, a residual U-block is employed, thereby forming an enhanced U-shaped network structure. In the experiments, we employ HD95, Dice, and Recall as evaluation metrics. MPANet demonstrates superior performance compared to state-of-the-art networks on three our self-collected datasets, showing improvements of over 1% across all metrics. In addition, we validate the efficacy of MPA as a plug-and-play module and the benefits of applying deep supervision to multidecoder network.

Accumulating evidence indicates that microRNAs (miRNAs) can control and coordinate various biological processes. Consequently, abnormal expressions of miRNAs have been linked to various complex diseases. Recognizable proof of miRNA-disease associations (MDAs) will contribute to the diagnosis and treatment of human diseases. Nevertheless, traditional experimental verification of MDAs is laborious and limited to small-scale. Therefore, it is necessary to develop reliable and effective computational methods to predict novel MDAs. In this work, a multi-kernel graph attention deep autoencoder (MGADAE) method is proposed to predict potential MDAs. In detail, MGADAE first employs the multiple kernel learning (MKL) algorithm to construct an integrated miRNA similarity and disease similarity, providing more biological information for further feature learning. Second, MGADAE combines the known MDAs, disease similarity, and miRNA similarity into a heterogeneous network, then learns the representations of miRNAs and diseases through graph convolution operation. After that, an attention mechanism is introduced into MGADAE to integrate the representations from multiple graph convolutional network (GCN) layers. Lastly, the integrated representations of miRNAs and diseases are input into the bilinear decoder to obtain the final predicted association scores. Corresponding experiments prove that the proposed method outperforms existing advanced approaches in MDA prediction. Furthermore, case studies related to two human cancers provide further confirmation of the reliability of MGADAE in practice.

Synergistic drug combination prediction tasks based on the computational models have been widely studied and applied in the cancer field. However, most of models only consider the interactions between drug pairs and specific cell lines, without taking into account the multiple biological relationships of drug-drug and cell line-cell line that also largely affect synergistic mechanisms. To this end, here we propose a multi-modal deep learning framework, termed MDNNSyn, which adequately applies multi-source information and trains multi-modal features to infer potential synergistic drug combinations. MDNNSyn extracts topology modality features by implementing the multi-layer hypergraph neural network on drug synergy hypergraph and constructs semantic modality features through similarity strategy. A multi-modal fusion network layer with gated neural network is then employed for synergy score prediction. MDNNSyn is compared to five classic and state-of-the-art prediction methods on DrugCombDB and Oncology-Screen datasets. The model achieves area under the curve (AUC) scores of 0.8682 and 0.9013 on two datasets, an improvement of 3.70 % and 2.71 % over the second-best model. Case study indicates that MDNNSyn is capable of detecting potential synergistic drug combinations.

Recent methods often introduce attention mechanisms into the skip connections of U-shaped networks to capture features. However, these methods usually overlook spatial information extraction in skip connections and exhibit inefficiency in capturing spatial and channel information. This issue prompts us to reevaluate the design of the skip-connection mechanism and propose a new deep-learning network called the Fusing Spatial and Channel Attention Network, abbreviated as FSCA-Net. FSCA-Net is a novel U-shaped network architecture that utilizes the Parallel Attention Transformer (PAT) to enhance the extraction of spatial and channel features in the skip-connection mechanism, further compensating for downsampling losses. We design the Cross-Attention Bridge Layer (CAB) to mitigate excessive feature and resolution loss when downsampling to the lowest level, ensuring meaningful information fusion during upsampling at the lowest level. Finally, we construct the Dual-Path Channel Attention (DPCA) module to guide channel and spatial information filtering for Transformer features, eliminating ambiguities with decoder features and better concatenating features with semantic inconsistencies between the Transformer and the U-Net decoder. FSCA-Net is designed explicitly for fine-grained segmentation tasks of multiple organs and regions. Our approach achieves over 48% reduction in FLOPs and over 32% reduction in parameters compared to the state-of-the-art method. Moreover, FSCA-Net outperforms existing segmentation methods on seven public datasets, demonstrating exceptional performance.

The pathogenesis of Alzheimer's disease (AD) is extremely intricate, which makes AD patients almost incurable. Recent studies have demonstrated that analyzing multi-modal data can offer a comprehensive perspective on the different stages of AD progression, which is beneficial for early diagnosis of AD. In this paper, we propose a deep self-reconstruction fusion similarity hashing (DS-FSH) method to effectively capture the AD-related biomarkers from the multi-modal data and leverage them to diagnose AD. Given that most existing methods ignore the topological structure of the data, a deep self-reconstruction model based on random walk graph regularization is designed to reconstruct the multi-modal data, thereby learning the nonlinear relationship between samples. Additionally, a fused similarity hash based on anchor graph is proposed to generate discriminative binary hash codes for multi-modal reconstructed data. This allows sample fused similarity to be effectively modeled by a fusion similarity matrix based on anchor graph while modal correlation can be approximated by Hamming distance. Especially, extracted features from the multi-modal data are classified using deep sparse autoencoders classifier. Finally, experiments conduct on the AD Neuroimaging Initiative database show that DS-FSH outperforms comparable methods of AD classification. To conclude, DS-FSH identifies multi-modal features closely associated with AD, which are expected to contribute significantly to understanding of the pathogenesis of AD.

The Transformer-based methods provide a good opportunity for modeling the global context of gigapixel whole slide image (WSI), however, there are still two main problems in applying Transformer to WSI-based survival analysis task. First, the training data for survival analysis is limited, which makes the model prone to overfitting. This problem is even worse for Transformer-based models which require large-scale data to train. Second, WSI is of extremely high resolution (up to 150,000 x 150,000 pixels) and is typically organized as a multi-resolution pyramid. Vanilla Transformer cannot model the hierarchical structure of WSI (such as patch cluster-level relationships), which makes it incapable of learning hierarchical WSI representation. To address these problems, in this paper, we propose a novel Sparse and Hierarchical Transformer (SH-Transformer) for survival analysis. Specifically, we introduce sparse self-attention to alleviate the overfitting problem, and propose a hierarchical Transformer structure to learn the hierarchical WSI representation. Experimental results based on three WSI datasets show that the proposed framework outperforms the state-of-the-art methods.

Single-cell RNA sequencing (scRNA-seq) technology is famous for providing a microscopic view to help capture cellular heterogeneity. This characteristic has advanced the field of genomics by enabling the delicate differentiation of cell types. However, the properties of single-cell datasets, such as high dropout events, noise, and high dimensionality, are still a research challenge in the single-cell field. To utilize single-cell data more efficiently and to better explore the heterogeneity among cells, a new graph autoencoder (GAE)-based consensus-guided model (scGAC) is proposed in this article. The data are preprocessed into multiple top-level feature datasets. Then, feature learning is performed by using GAEs to generate new feature matrices, followed by similarity learning based on distance fusion methods. The learned similarity matrices are fed back to the GAEs to guide their feature learning process. Finally, the abovementioned steps are iterated continuously to integrate the final consistent similarity matrix and perform other related downstream analyses. The scGAC model can accurately identify critical features and effectively preserve the internal structure of the data. This can further improve the accuracy of cell type identification.

CircRNA as a biomarker has been shown to have an essential effect on the occurrence and prognosis of a wide range of human diseases. Because of the high cost of wet experiments, computational methods are widely used to explore circRNA. However, the performance and robustness of the computational models still need to be further improved. To solve these problems, this paper proposes a novel method based on graph random propagation network and multi-head dynamic graph attention network (GRPGAT) to predict the potential associations between circRNAs and diseases. Firstly, GRPGAT uses centered kernel alignment method to fuse the circRNA similarity kernels and disease similarity kernels. Then the integrated vectors build a heterogeneous graph and are sent to a graph random propagation network. The remaining nodes are fed into a multi-head dynamic attention network for feature extraction. Finally, a four-layer Multilayer Perceptron is used to learn features and gain the prediction scores. Experiments are supported by cirR2Disease, and achieve Area Under Curve (AUC) scores of 0.9636 in 5-fold cross validation. In comparison with the state-of-the-art models, GRPGAT also shows superior performance.

Antiviral peptides (AVPs) are widely found in animals and plants, with high specificity and strong sensitivity to drug-resistant viruses. However, due to the great heterogeneity of different viruses, most of the AVPs have specific antiviral activities. Therefore, it is necessary to identify the specific activities of AVPs on virus types. Most existing studies only identify AVPs, with only a few studies identifying subclasses by training multiple binary classifiers. We develop a two-stage prediction tool named FFMAVP that can simultaneously predict AVPs and their subclasses. In the first stage, we identify whether a peptide is AVP or not. In the second stage, we predict the six virus families and eight species specifically targeted by AVPs based on two multiclass tasks. Specifically, the feature extraction module in the two-stage task of FFMAVP adopts the same neural network structure, in which one branch extracts features based on amino acid feature descriptors and the other branch extracts sequence features. Then, the two types of features are fused for the following task. Considering the correlation between the two tasks of the second stage, a multitask learning model is constructed to improve the effectiveness of the two multiclass tasks. In addition, to improve the effectiveness of the second stage, the network parameters trained through the first-stage data are used to initialize the network parameters in the second stage. As a demonstration, the cross-validation results, independent test results and visualization results show that FFMAVP achieves great advantages in both stages.

A large number of works have presented the single-cell RNA sequencing (scRNA-seq) to study the diversity and biological functions of cells at the single-cell level. Clustering identifies unknown cell types, which is essential for downstream analysis of scRNA-seq samples. However, the high dimensionality, high noise and pervasive dropout rate of scRNA-seq samples have a significant challenge to the cluster analysis of scRNA-seq samples. Herein, we propose a new adaptive fuzzy clustering model based on the denoising autoencoder and self-attention mechanism called the scDASFK. It implements the comparative learning to integrate cell similar information into the clustering method and uses a deep denoising network module to denoise the data. scDASFK consists of a self-attention mechanism for further denoising where an adaptive clustering optimization function for iterative clustering is implemented. In order to make the denoised latent features better reflect the cell structure, we introduce a new adaptive feedback mechanism to supervise the denoising process through the clustering results. Experiments on 16 real scRNA-seq datasets show that scDASFK performs well in terms of clustering accuracy, scalability and stability. Overall, scDASFK is an effective clustering model with great potential for scRNA-seq samples analysis. Our scDASFK model codes are freely available at https://github.com/LRX2022/scDASFK.

The advances in single-cell ribonucleic acid sequencing (scRNA-seq) allow researchers to explore cellular heterogeneity and human diseases at cell resolution. Cell clustering is a prerequisite in scRNA-seq analysis since it can recognize cell identities. However, the high dimensionality, noises and significant sparsity of scRNA-seq data have made it a big challenge. Although many methods have emerged, they still fail to fully explore the intrinsic properties of cells and the relationship among cells, which seriously affects the downstream clustering performance. Here, we propose a new deep contrastive clustering algorithm called scDCCA. It integrates a denoising auto-encoder and a dual contrastive learning module into a deep clustering framework to extract valuable features and realize cell clustering. Specifically, to better characterize and learn data representations robustly, scDCCA utilizes a denoising Zero-Inflated Negative Binomial model-based auto-encoder to extract low-dimensional features. Meanwhile, scDCCA incorporates a dual contrastive learning module to capture the pairwise proximity of cells. By increasing the similarities between positive pairs and the differences between negative ones, the contrasts at both the instance and the cluster level help the model learn more discriminative features and achieve better cell segregation. Furthermore, scDCCA joins feature learning with clustering, which realizes representation learning and cell clustering in an end-to-end manner. Experimental results of 14 real datasets validate that scDCCA outperforms eight state-of-the-art methods in terms of accuracy, generalizability, scalability and efficiency. Cell visualization and biological analysis demonstrate that scDCCA significantly improves clustering and facilitates downstream analysis for scRNA-seq data. The code is available at https://github.com/WJ319/scDCCA.

Although previous studies have revealed that synonymous mutations contribute to various human diseases, distinguishing deleterious synonymous mutations from benign ones is still a challenge in medical genomics. Recently, computational tools have been introduced to predict the harmfulness of synonymous mutations. However, most of these computational tools rely on balanced training sets without considering abundant negative samples that could result in deficient performance. In this study, we propose a computational model that uses a selective ensemble to predict deleterious synonymous mutations (seDSM). We construct several candidate base classifiers for the ensemble using balanced training subsets randomly sampled from the imbalanced benchmark training sets. The diversity measures of the base classifiers are calculated by the pairwise diversity metrics, and the classifiers with the highest diversities are selected for integration using soft voting for synonymous mutation prediction. We also design two strategies for filling in missing values in the imbalanced dataset and constructing models using different pairwise diversity metrics. The experimental results show that a selective ensemble based on double fault with the ensemble strategy EKNNI for filling in missing values is the most effective scheme. Finally, using 40-dimensional biology features, we propose a novel model based on a selective ensemble for predicting deleterious synonymous mutations (seDSM). seDSM outperformed other state-of-the-art methods on the independent test sets according to multiple evaluation indicators, indicating that it has an outstanding predictive performance for deleterious synonymous mutations. We hope that seDSM will be useful for studying deleterious synonymous mutations and advancing our understanding of synonymous mutations. The source code of seDSM is freely accessible at https://github.com/xialab-ahu/seDSM.git.

The progress of single-cell RNA sequencing (scRNA-seq) has led to a large number of scRNA-seq data, which are widely used in biomedical research. The noise in the raw data and tens of thousands of genes pose a challenge to capture the real structure and effective information of scRNA-seq data. Most of the existing single-cell analysis methods assume that the low-dimensional embedding of the raw data belongs to a Gaussian distribution or a low-dimensional nonlinear space without any prior information, which limits the flexibility and controllability of the model to a great extent. In addition, many existing methods need high computational cost, which makes them difficult to be used to deal with large-scale datasets. Here, we design and develop a depth generation model named Gaussian mixture adversarial autoencoders (scGMAAE), assuming that the low-dimensional embedding of different types of cells follows different Gaussian distributions, integrating Bayesian variational inference and adversarial training, as to give the interpretable latent representation of complex data and discover the statistical distribution of different types of cells. The scGMAAE is provided with good controllability, interpretability and scalability. Therefore, it can process large-scale datasets in a short time and give competitive results. scGMAAE outperforms existing methods in several ways, including dimensionality reduction visualization, cell clustering, differential expression analysis and batch effect removal. Importantly, compared with most deep learning methods, scGMAAE requires less iterations to generate the best results.

BACKGROUND: Single-cell RNA sequencing (scRNA-seq) strives to capture cellular diversity with higher resolution than bulk RNA sequencing. Clustering analysis is critical to transcriptome research as it allows for further identification and discovery of new cell types. Unsupervised clustering cannot integrate prior knowledge where relevant information is widely available. Purely unsupervised clustering algorithms may not yield biologically interpretable clusters when confronted with the high dimensionality of scRNA-seq data and frequent dropout events, which makes identification of cell types more challenging. RESULTS: We propose scSemiAAE, a semi-supervised clustering model for scRNA sequence analysis using deep generative neural networks. Specifically, scSemiAAE carefully designs a ZINB adversarial autoencoder-based architecture that inherently integrates adversarial training and semi-supervised modules in the latent space. In a series of experiments on scRNA-seq datasets spanning thousands to tens of thousands of cells, scSemiAAE can significantly improve clustering performance compared to dozens of unsupervised and semi-supervised algorithms, promoting clustering and interpretability of downstream analyses. CONCLUSION: scSemiAAE is a Python-based algorithm implemented on the VSCode platform that provides efficient visualization, clustering, and cell type assignment for scRNA-seq data. The tool is available from https://github.com/WHang98/scSemiAAE .

Accumulating evidence suggests that circRNAs play crucial roles in human diseases. CircRNA-disease association prediction is extremely helpful in understanding pathogenesis, diagnosis, and prevention, as well as identifying relevant biomarkers. During the past few years, a large number of deep learning (DL) based methods have been proposed for predicting circRNA-disease association and achieved impressive prediction performance. However, there are two main drawbacks to these methods. The first is these methods underutilize biometric information in the data. Second, the features extracted by these methods are not outstanding to represent association characteristics between circRNAs and diseases. In this study, we developed a novel deep learning model, named iCircDA-NEAE, to predict circRNA-disease associations. In particular, we use disease semantic similarity, Gaussian interaction profile kernel, circRNA expression profile similarity, and Jaccard similarity simultaneously for the first time, and extract hidden features based on accelerated attribute network embedding (AANE) and dynamic convolutional autoencoder (DCAE). Experimental results on the circR2Disease dataset show that iCircDA-NEAE outperforms other competing methods significantly. Besides, 16 of the top 20 circRNA-disease pairs with the highest prediction scores were validated by relevant literature. Furthermore, we observe that iCircDA-NEAE can effectively predict new potential circRNA-disease associations.

Due to the complexity of cancer pathogenesis at different omics levels, it is necessary to find a comprehensive method to accurately distinguish and find cancer subtypes for cancer treatment. In this paper, we proposed a new cancer multi-omics subtype identification method, which is based on variational autoencoder measured by Wasserstein distance and graph autoencoder (WVGMO). This method depends on two foremost models. The first model is a variational autoencoder measured by Wasserstein distance (WVAE), which is used to extract potential spatial information of each omic data type. The second model is the graph autoencoder (GAE) with the second-order proximity. It has the capability to retain the topological structure information and feature information of the multi-omics data. And then, the identification of cancer subtypes via k-means clustering. Extensive experiments were conducted on seven different cancers based on four omics data from TCGA. The results show that WVGMO provides equivalent or even better results than the most of advanced synthesis methods.

Due to the global outbreak of COVID-19 and its variants, antiviral peptides with anti-coronavirus activity (ACVPs) represent a promising new drug candidate for the treatment of coronavirus infection. At present, several computational tools have been developed to identify ACVPs, but the overall prediction performance is still not enough to meet the actual therapeutic application. In this study, we constructed an efficient and reliable prediction model PACVP (Prediction of Anti-CoronaVirus Peptides) for identifying ACVPs based on effective feature representation and a two-layer stacking learning framework. In the first layer, we use nine feature encoding methods with different feature representation angles to characterize the rich sequence information and fuse them into a feature matrix. Secondly, data normalization and unbalanced data processing are carried out. Next, 12 baseline models are constructed by combining three feature selection methods and four machine learning classification algorithms. In the second layer, we input the optimal probability features into the logistic regression algorithm (LR) to train the final model PACVP. The experiments show that PACVP achieves favorable prediction performance on independent test dataset, with ACC of 0.9208 and AUC of 0.9465. We hope that PACVP will become a useful method for identifying, annotating and characterizing novel ACVPs.

Biclustering algorithms are essential for processing gene expression data. However, to process the dataset, most biclustering algorithms require preprocessing the data matrix into a binary matrix. Regrettably, this type of preprocessing may introduce noise or cause information loss in the binary matrix, which would reduce the biclustering algorithm's ability to effectively obtain the optimal biclusters. In this paper, we propose a new preprocessing method named Mean-Standard Deviation (MSD) to resolve the problem. Additionally, we introduce a new biclustering algorithm called Weight Adjacency Difference Matrix Binary Biclustering (W-AMBB) to effectively process datasets containing overlapping biclusters. The basic idea is to create a weighted adjacency difference matrix by applying weights to a binary matrix that is derived from the data matrix. This allows us to identify genes with significant associations in sample data by efficiently identifying similar genes that respond to specific conditions. Furthermore, the performance of the W-AMBB algorithm was tested on both synthetic and real datasets and compared with other classical biclustering methods. The experiment results demonstrate that the W-AMBB algorithm is significantly more robust than the compared biclustering methods on the synthetic dataset. Additionally, the results of the GO enrichment analysis show that the W-AMBB method possesses biological significance on real datasets.

Gene regulatory networks (GRNs) participate in many biological processes, and reconstructing them plays an important role in systems biology. Although many advanced methods have been proposed for GRN reconstruction, their predictive performance is far from the ideal standard, so it is urgent to design a more effective method to reconstruct GRN. Moreover, most methods only consider the gene expression data, ignoring the network structure information contained in GRN. In this study, we propose a supervised model named CNNGRN, which infers GRN from bulk time-series expression data via convolutional neural network (CNN) model, with a more informative feature. Bulk time series gene expression data imply the intricate regulatory associations between genes, and the network structure feature of ground-truth GRN contains rich neighbor information. Hence, CNNGRN integrates the above two features as model inputs. In addition, CNN is adopted to extract intricate features of genes and infer the potential associations between regulators and target genes. Moreover, feature importance visualization experiments are implemented to seek the key features. Experimental results show that CNNGRN achieved competitive performance on benchmark datasets compared to the state-of-the-art computational methods. Finally, hub genes identified based on CNNGRN have been confirmed to be involved in biological processes through literature.

Single-cell RNA sequencing technology provides powerful support for researchers to understand the complex mechanisms of cells at the single-cell level. Due to the high sparsity, technical noise, and computational complexity of single-cell transcriptome data, the existing data analysis methods are unable to effectively extract the fine-grained characteristics of scRNA-seq data, resulting in inaccurately analyze the heterogeneity of the individual cell from a great quantity of cell mixtures. To address these shortcomings, we proposed an end-to-end analysis method called dhaSCA, which integrates the Graph convolutional neural network (GCN) feature learning and downstream tasks such as classification and imputation into a unified deep learning manner. dhaSCA uses hybrid GCN-MLP deep autoencoder and to capture structural information between cells, and learn the low dimensional cell representation. It also introduces downstream tasks as constraints to guide the model to learn more accurate cell features. We conducted various experiments to evaluate the performance of dhaSCA based on eight real RNA-Seq datasets, including classification, imputation, clustering, and visualization. The results show that dhaSCA outperforms other state-of-the-art methods in these downstream tasks. Therefore, dhaSCA is able to obtain a richer representation of cells, and provides strong support for efficient analysis of single-cell data.

CircRNAs have a stable structure, which gives them a higher tolerance to nucleases. Therefore, the properties of circular RNAs are beneficial in disease diagnosis. However, there are few known associations between circRNAs and disease. Biological experiments identify new associations is time-consuming and high-cost. As a result, there is a need of building efficient and achievable computation models to predict potential circRNA-disease associations. In this paper, we design a novel convolution neural networks framework(DMFCNNCD) to learn features from deep matrix factorization to predict circRNA-disease associations. Firstly, we decompose the circRNA-disease association matrix to obtain the original features of the disease and circRNA, and use the mapping module to extract potential nonlinear features. Then, we integrate it with the similarity information to form a training set. Finally, we apply convolution neural networks to predict the unknown association between circRNAs and diseases. The five-fold cross-validation on various experiments shows that our method can predict circRNA-disease association and outperforms state of the art methods.

An Increase in microbial activity is shown to be intimately connected with the pathogenesis of diseases. Considering the expense of traditional verification methods, researchers are working to develop high-efficiency methods for detecting potential disease-related microbes. In this article, a new prediction method, MSF-LRR, is established, which uses Low-Rank Representation (LRR) to perform multi-similarity information fusion to predict disease-related microbes. Considering that most existing methods only use one class of similarity, three classes of microbe and disease similarity are added. Then, LRR is used to obtain low-rank structural similarity information. Additionally, the method adaptively extracts the local low-rank structure of the data from a global perspective, to make the information used for the prediction more effective. Finally, a neighbor-based prediction method that utilizes the concept of collaborative filtering is applied to predict unknown microbe-disease pairs. As a result, the AUC value of MSF-LRR is superior to other existing algorithms under 5-fold cross-validation. Furthermore, in case studies, excluding originally known associations, 16 and 19 of the top 20 microbes associated with Bacterial Vaginosis and Irritable Bowel Syndrome, respectively, have been confirmed by the recent literature. In summary, MSF-LRR is a good predictor of potential microbe-disease associations and can contribute to drug discovery and biological research.

Single-cell RNA sequencing (scRNA-Seq) technology has emerged as a powerful tool to investigate cellular heterogeneity within tissues, organs, and organisms. One fundamental question pertaining to single-cell gene expression data analysis revolves around the identification of cell types, which constitutes a critical step within the data processing workflow. However, existing methods for cell type identification through learning low-dimensional latent embeddings often overlook the intercellular structural relationships. In this paper, we present a novel non-negative low-rank similarity correction model (NLRSIM) that leverages subspace clustering to preserve the global structure among cells. This model introduces a novel manifold learning process to address the issue of imbalanced neighbourhood spatial density in cells, thereby effectively preserving local geometric structures. This procedure utilizes a position-sensitive hashing algorithm to construct the graph structure of the data. The experimental results demonstrate that the NLRSIM surpasses other advanced models in terms of clustering effects and visualization experiments. The validated effectiveness of gene expression information after calibration by the NLRSIM model has been duly ascertained in the realm of relevant biological studies. The NLRSIM model offers unprecedented insights into gene expression, states, and structures at the individual cellular level, thereby contributing novel perspectives to the field.

Circular RNAs (circRNAs) are a category of noncoding RNAs that exist in great numbers in eukaryotes. They have recently been discovered to be crucial in the growth of tumors. Therefore, it is important to explore the association of circRNAs with disease. This paper proposes a new method based on DeepWalk and nonnegative matrix factorization (DWNMF) to predict circRNA-disease association. Based on the known circRNA-disease association, we calculate the topological similarity of circRNA and disease via the DeepWalk-based method to learn the node features on the association network. Next, the functional similarity of the circRNAs and the semantic similarity of the diseases are fused with their respective topological similarities at different scales. Then, we use the improved weightedK-nearest neighbor (IWKNN) method to preprocess the circRNA-disease association network and correct nonnegative associations by setting different parametersK1andK2in the circRNA and disease matrices. Finally, theL2,1-norm, dual-graph regularization term and Frobenius norm regularization term are introduced into the nonnegative matrix factorization model to predict the circRNA-disease correlation. We perform cross-validation on circR2Disease, circRNADisease, and MNDR. The numerical results show that DWNMF is an efficient tool for forecasting potential circRNA-disease relationships, outperforming other state-of-the-art approaches in terms of predictive performance.

With the discovery of causality between synonymous mutations and diseases, it has become increasingly important to identify deleterious synonymous mutations for better understanding of their functional mechanisms. Although several machine learning methods have been proposed to solve the task, an effective feature representation method that can make use of the inner difference and relevance between deleterious and benign synonymous mutations is still challenging considering the vast number of synonymous mutations in human genome. In this work, we developed a robust and accurate predictor called frDSM for deleterious synonymous mutation prediction using logistic regression. More specifically, we introduced an effective feature representation learning method which exploits multiple feature descriptors from different perspectives including functional scores obtained from previously computational methods, evolutionary conservation, splicing and sequence feature descriptors, and these features descriptors were input into the 76 XGBoost classifiers to obtain the predictive probabilities values. These probabilities were concatenated to generate the 76-dimension new feature vector, and feature selection method was used to remove redundant and irrelevant features. Experimental results show that frDSM enables robust and accurate prediction than the competing prediction methods with 31 optimal features, which demonstrated the effectiveness of the feature representation learning method. frDSM is freely available at http://frdsm.xialab.info.

With the development of bioinformatics, the important role played by lncRNAs in various intractable diseases has aroused the interest of many experts. In recent studies, researchers have found that several human diseases are related to lncRANs. Moreover, it is very difficult and expensive to explore the unknown lncRNA-disease associations (LDAs), so only a few associations have been confirmed. It is vital to find a more accurate and effective method to identify potential LDAs. In this study, a method of collaborative matrix factorization based on correntropy (LDCMFC) is proposed for the identification of potential LDAs. To improve the robustness of the algorithm, the traditional minimization of the Euclidean distance is replaced with the maximized correntropy. In addition, the weighted K nearest known neighbor (WKNKN) method is used to rebuild the adjacency matrix. Finally, the performance of LDCMFC is tested by 5-fold cross-validation. Compared with other traditional methods, LDACMFC obtains a higher AUC of 0.8628. In different types of studies of three important cancer cases, most of the potentially relevant lncRNAs derived from the experiments have been validated in the databases. The final result shows that LDCMFC is a feasible method to predict LDAs.

Hyperspectral image classification methods based on the graph convolutional network (GCN) have received more attention because they can handle irregular regions by graph encoding techniques. However, GCN-based HSI classification methods are highly sensitive to the quality of the graph structure. Its performance degrades in the case of underdeveloped graphs because it cannot excavate the intrinsic adjacency relationships. Thus, it is necessary to improve the quality of graph structure in GCN-based methods. In this paper, a novel diversity-connected graph convolutional network (DCGCN) method is proposed to improve the quality of the graph structure for HSI classification, and its basic idea can be adopted by other GCN-based methods. First, the potential neighbors are excavated by performing topological extensions based on the given graph. The diversity of surrounding neighbors is maintained by adaptively smoothing operation via a global threshold value from Kullback-Leibler divergence to eliminate weak interclass connections caused by weakly spectral variability. Second, another key connectivity restriction is imposed on the diverse neighbors to further refine the ambiguous connections of hard samples aiming at removing strong interclass connections where the spectral information is heavily confounded. Finally, the DCGCN method is analyzed theoretically to demonstrate its low-pass filter property. The comprehensive experiments demonstrate the effectiveness of the proposed DCGCN method and the basic idea of the diversity-connected graph in terms of overall accuracy (OA), kappa coefficient (KC), average accuracy (AA) indexes.

Single-cell RNA sequencing (scRNA-seq) technology can provide expression profile of single cells, which propels biological research into a new chapter. Clustering individual cells based on their transcriptome is a critical objective of scRNA-seq data analysis. However, the high-dimensional, sparse and noisy nature of scRNA-seq data pose a challenge to single-cell clustering. Therefore, it is urgent to develop a clustering method targeting scRNA-seq data characteristics. Due to its powerful subspace learning capability and robustness to noise, the subspace segmentation method based on low-rank representation (LRR) is broadly used in clustering researches and achieves satisfactory results. In view of this, we propose a personalized low-rank subspace clustering method, namely PLRLS, to learn more accurate subspace structures from both global and local perspectives. Specifically, we first introduce the local structure constraint to capture the local structure information of the data, while helping our method to obtain better inter-cluster separability and intra-cluster compactness. Then, in order to retain the important similarity information that is ignored by the LRR model, we utilize the fractional function to extract similarity information between cells, and introduce this information as the similarity constraint into the LRR framework. The fractional function is an efficient similarity measure designed for scRNA-seq data, which has theoretical and practical implications. In the end, based on the LRR matrix learned from PLRLS, we perform downstream analyses on real scRNA-seq datasets, including spectral clustering, visualization and marker gene identification. Comparative experiments show that the proposed method achieves superior clustering accuracy and robustness.

Single-cell RNA sequencing (scRNA-seq) has rapidly emerged as a powerful technique for analyzing cellular heterogeneity at the individual cell level. In the analysis of scRNA-seq data, cell clustering is a critical step in downstream analysis, as it enables the identification of cell types and the discovery of novel cell subtypes. However, the characteristics of scRNA-seq data, such as high dimensionality and sparsity, dropout events and batch effects, present significant computational challenges for clustering analysis. In this study, we propose scGCC, a novel graph self-supervised contrastive learning model, to address the challenges faced in scRNA-seq data analysis. scGCC comprises two main components: a representation learning module and a clustering module. The scRNA-seq data is first fed into a representation learning module for training, which is then used for data classification through a clustering module. scGCC can learn low-dimensional denoised embeddings, which is advantageous for our clustering task. We introduce Graph Attention Networks (GAT) for cell representation learning, which enables better feature extraction and improved clustering accuracy. Additionally, we propose five data augmentation methods to improve clustering performance by increasing data diversity and reducing overfitting. These methods enhance the robustness of clustering results. Our experimental study on 14 real-world datasets has demonstrated that our model achieves extraordinary accuracy and robustness. We also perform downstream tasks, including batch effect removal, trajectory inference, and marker genes analysis, to verify the biological effectiveness of our model.

Studies have revealed that microbes have an important effect on numerous physiological processes, and further research on the links between diseases and microbes is significant. Given that laboratory methods are expensive and not optimized, computational models are increasingly used for discovering disease-related microbes. Here, a new neighbor approach based on two-tier Bi-Random Walk is proposed for potential disease-related microbes, known as NTBiRW. In this method, the first step is to construct multiple microbe similarities and disease similarities. Then, three kinds of microbe/disease similarity are integrated through two-tier Bi-Random Walk to obtain the final integrated microbe/disease similarity network with different weights. Finally, Weighted K Nearest Known Neighbors (WKNKN) is used for prediction based on the final similarity network. In addition, leave-one-out cross-validation (LOOCV) and 5-fold cross-validation (5-fold CV) are applied for evaluating the performance of NTBiRW. Multiple evaluating indicators are taken to show the performance from multiple perspectives. And most of the evaluation index values of NTBiRW are better than those of the compared methods. Moreover, in case studies on atopic dermatitis and psoriasis, most of the first 10 candidates in the final result can be proven. This also demonstrates the capability of NTBiRW for discovering new associations. Therefore, this method can contribute to the discovery of disease-related microbes and thus offer new thoughts for further understanding the pathogenesis of diseases.

Determining microRNA (miRNA)-disease associations (MDAs) is an integral part in the prevention, diagnosis, and treatment of complex diseases. However, wet experiments to discern MDAs are inefficient and expensive. Hence, the development of reliable and efficient data integrative models for predicting MDAs is of significant meaning. In the present work, a novel deep learning method for predicting MDAs through deep autoencoder with multiple kernel learning (DAEMKL) is presented. Above all, DAEMKL applies multiple kernel learning (MKL) in miRNA space and disease space to construct miRNA similarity network and disease similarity network, respectively. Then, for each disease or miRNA, its feature representation is learned from the miRNA similarity network and disease similarity network via the regression model. After that, the integrated miRNA feature representation and disease feature representation are input into deep autoencoder (DAE). Furthermore, the novel MDAs are predicted through reconstruction error. Ultimately, the AUC results show that DAEMKL achieves outstanding performance. In addition, case studies of three complex diseases further prove that DAEMKL has excellent predictive performance and can discover a large number of underlying MDAs. On the whole, our method DAEMKL is an effective method to identify MDAs.

Immunotherapy is an excellent treatment option for many solid tumors, and the therapeutic effect has been proved in clinical. Microsatellite instability (MSI) has been an important predictive marker for response to immune checkpoint inhibitors for Colorectal cancer (CRC). CRC patients with high-level MSI can be provided immunotherapy and benefit from it. Some studies have attempted to predict MSI using pathological images based on deep learning, but the accuracy of the prediction model needs to be improved. In this study, considering that different scales of pathological images contain various levels of information, we propose a novel method, named MSIUMP, to predict microsatellite instability using multi-scale pathological images. We first predict the MSI of a patient based on deep learning model using different scale pathological images, and then integrate the results by ensemble learning to improve the generalization and robustness. In addition, a convolutional neural network model modified on the basis of EfficientNet is used to extract the information of patches from pathological images at different scales. Our method achieved the areas under the receiver operating characteristic curves (AUC) of 0.9096 in the internal test dataset TCGA-CRC and the AUC of 0.9619 in the external independent validation dataset PAIP2020. These results demonstrate the potential of our method as a prediction tool for microsatellite instability in colorectal cancer.

Neuroimaging genetics is a powerful technique for discovering the relationships between genotype and imaging phenotype. However, many univariate or multivariate regression approaches have only focused on imaging quantitative traits (QTs) that are relevant to some genetic markers on distinct pathways and might not be disease specific. In addition, there are complex relations between subjects of distinct modalities and diagnosis labels, which contain useful information for the treatment of Alzheimer’s disease (AD). Here, a novel probability connectivity-based penalty is developed for incorporating the prior information to explore relations among different subjects with disease status. Specifically, the Pearson’s correlation coefficient (PCC) is used to construct a similarity matrix in a probability graph, first to express the connectivity weights between subjects, which can reflect the different correlations among subjects within the same class. Second, a diagnosis-aligned probability connectivity-based multimodal regression (DPCMR) method is employed to find the relations among modalities of distinct subjects. It also mines associations between genetic markers and imaging phenotypes. The AD risk single nucleotide polymorphism (SNP) APOE rs429358 and three modalities of neuroimaging data are used to verify the performance of all of the methods. The experimental results reveal that DPCMR has better performance and identifies some brain regions across multiple modalities related to diseases.

Ensemble learning is to train and combine multiple learners to complete the corresponding learning tasks. It can improve the stability of the overall model, and a good ensemble method can further improve the accuracy of the model. At the same time, as one of the outstanding representatives of machine learning, Extreme Learning Machine has attracted the continuous attention of experts and scholars. to get a better representation of the feature space, we extend the Gaussian kernel in the kernel risk-sensitive loss and propose a Kernel Risk-Sensitive Loss of q-Gaussian kernel and Hyper-graph Regularized Extreme Learning Machine method. Since the contingency in the ELM training process cannot be completely avoided, the stability of most ELM methods is affected to some extent. What’s more, we introduce the voting mechanism and a new ELM classification model named Kernel Risk-Sensitive Loss of q-Gaussian kernel and Hyper-graph Regularized Integrated Extreme Learning Machine based on Voting Mechanism is proposed. It improves the stability of the model through the idea of ensemble learning. We apply the new model on six real data sets, and through observation and analysis of experimental results, we find that the new model has certain competitiveness, especially in classification accuracy and stability.

A smart recommendation method can greatly improve tourists' travel experience, and it is an important task for tourism recommendation systems to intelligently suggest scenic spots for tourists according to their historical visit records. Currently, the collaborative filtering and deep neural network-based methods occupy the mainstream of tourism recommendation systems. Although each type of recommendation methods is superior over the others in terms of different aspects, the performance of a single recommendation method is limited. In order to inherit the advantages of different types of recommendation methods, this work suggests a hybrid method for assembling multiple methods for tourism recommendation. Based on the scenic spots obtained by multiple recommendation methods, the proposed hybrid method uses two novel objectives to evaluate each scenic spot, and identifies the best$K$scenic spots via the techniques used in evolutionary multi-objective optimization. In comparison to existing recommendation methods and hybrid methods, the proposed hybrid method exhibits better performance on two public tourism datasets and a new dataset created based on the tourism information of Huangshan City.

Neuropeptides (NPs) are a particular class of informative substances in the immune system and physiological regulation. They play a crucial role in regulating physiological functions in various biological growth and developmental stages. In addition, NPs are crucial for developing new drugs for the treatment of neurological diseases. With the development of molecular biology techniques, some data-driven tools have emerged to predict NPs. However, it is necessary to improve the predictive performance of these tools for NPs. In this study, we developed a deep learning model (NeuroPred-CLQ) based on the temporal convolutional network (TCN) and multi-head attention mechanism to identify NPs effectively and translate the internal relationships of peptide sequences into numerical features by the Word2vec algorithm. The experimental results show that NeuroPred-CLQ learns data information effectively, achieving 93.6% accuracy and 98.8% AUC on the independent test set. The model has better performance in identifying NPs than the state-of-the-art predictors. Visualization of features using t-distribution random neighbor embedding shows that the NeuroPred-CLQ can clearly distinguish the positive NPs from the negative ones. We believe the NeuroPred-CLQ can facilitate drug development and clinical trial studies to treat neurological disorders.

With the development of research on the complex aetiology of many diseases, computational drug repositioning methodology has proven to be a shortcut to costly and inefficient traditional methods. Therefore, developing more promising computational methods is indispensable for finding new candidate diseases to treat with existing drugs. In this paper, a model integrating a new variant of message passing neural network and a novel-gated fusion mechanism called GLGMPNN is proposed for drug-disease association prediction. First, a light-gated message passing neural network (LGMPNN), including message passing, aggregation and updating, is proposed to separately extract multiple pieces of information from the similarity networks and the association network. Then, a gated fusion mechanism consisting of a forget gate and an output gate is applied to integrate the multiple pieces of information to extent. The forget gate calculated by the multiple embeddings is built to integrate the association information into the similarity information. Furthermore, the final node representations are controlled by the output gate, which fuses the topology information of the networks and the initial similarity information. Finally, a bilinear decoder is adopted to reconstruct an adjacency matrix for drug-disease associations. Evaluated by 10-fold cross-validations, GLGMPNN achieves excellent performance compared with the current models. The following studies show that our model can effectively discover novel drug-disease associations.

The rapid development of biomedicine has produced a large number of biomedical written materials. These unstructured text data create serious challenges for biomedical researchers to find information. Biomedical named entity recognition (BioNER) and biomedical relation extraction (BioRE) are the two most fundamental tasks of biomedical text mining. Accurately and efficiently identifying entities and extracting relations have become very important. Methods that perform two tasks separately are called pipeline models, and they have shortcomings such as insufficient interaction, low extraction quality and easy redundancy. To overcome the above shortcomings, many deep learning-based joint name entity recognition and relation extraction models have been proposed, and they have achieved advanced performance. This paper comprehensively summarize deep learning models for joint name entity recognition and relation extraction for biomedicine. The joint BioNER and BioRE models are discussed in the light of the challenges existing in the BioNER and BioRE tasks. Five joint BioNER and BioRE models and one pipeline model are selected for comparative experiments on four biomedical public datasets, and the experimental results are analyzed. Finally, we discuss the opportunities for future development of deep learning-based joint BioNER and BioRE models.

Rapid development of single-cell RNA sequencing (scRNA-seq) technology has allowed researchers to explore biological phenomena at the cellular scale. Clustering is a crucial and helpful step for researchers to study the heterogeneity of cell. Although many clustering methods have been proposed, massive dropout events and the curse of dimensionality in scRNA-seq data make it still difficult to analysis because they reduce the accuracy of clustering methods, leading to misidentification of cell types. In this work, we propose the scHFC, which is a hybrid fuzzy clustering method optimized by natural computation based on Fuzzy C Mean (FCM) and Gath-Geva (GG) algorithms. Specifically, principal component analysis algorithm is utilized to reduce the dimensions of scRNA-seq data after it is preprocessed. Then, FCM algorithm optimized by simulated annealing algorithm and genetic algorithm is applied to cluster the data to output a membership matrix, which represents the initial clustering result and is taken as the input for GG algorithm to get the final clustering results. We also develop a cluster number estimation method called multi-index comprehensive estimation, which can estimate the cluster numbers well by combining four clustering effectiveness indexes. The performance of the scHFC method is evaluated on 17 scRNA-seq datasets, and compared with six state-of-the-art methods. Experimental results validate the better performance of our scHFC method in terms of clustering accuracy and stability of algorithm. In short, scHFC is an effective method to cluster cells for scRNA-seq data, and it presents great potential for downstream analysis of scRNA-seq data. The source code is available at https://github.com/WJ319/scHFC.

MOTIVATION: A large number of studies have shown that clustering is a crucial step in scRNA-seq analysis. Most existing methods are based on unsupervised learning without the prior exploitation of any domain knowledge, which does not utilize available gold-standard labels. When confronted by the high dimensionality and general dropout events of scRNA-seq data, purely unsupervised clustering methods may not produce biologically interpretable clusters, which complicate cell type assignment. RESULTS: In this article, we propose a semi-supervised clustering method based on a capsule network named scCNC that integrates domain knowledge into the clustering step. Significantly, we also propose a Semi-supervised Greedy Iterative Training method used to train the whole network. Experiments on some real scRNA-seq datasets show that scCNC can significantly improve clustering performance and facilitate downstream analyses. AVAILABILITY AND IMPLEMENTATION: The source code of scCNC is freely available at https://github.com/WHY-17/scCNC. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Biclustering algorithm is an effective tool for processing gene expression datasets. There are two kinds of data matrices, binary data and non-binary data, which are processed by biclustering method. A binary matrix is usually converted from pre-processed gene expression data, which can effectively reduce the interference from noise and abnormal data, and is then processed using a biclustering algorithm. However, biclustering algorithms of dealing with binary data have a poor balance between running time and performance. In this paper, we propose a new biclustering algorithm called the Adjacency Difference Matrix Binary Biclustering algorithm (AMBB) for dealing with binary data to address the drawback. The AMBB algorithm constructs the adjacency matrix based on the adjacency difference values, and the submatrix obtained by continuously updating the adjacency difference matrix is called a bicluster. The adjacency matrix allows for clustering of gene that undergo similar reactions under different conditions into clusters, which is important for subsequent genes analysis. Meanwhile, experiments on synthetic and real datasets visually demonstrate that the AMBB algorithm has high practicability.

Enhancers are small regions of DNA that bind to proteins, which enhance the transcription of genes. The enhancer may be located upstream or downstream of the gene. It is not necessarily close to the gene to be acted on, because the entanglement structure of chromatin allows the positions far apart in the sequence to have the opportunity to contact each other. Therefore, identifying enhancers and their strength is a complex and challenging task. In this article, a new prediction method based on deep learning is proposed to identify enhancers and enhancer strength, called iEnhancer-DCLA. Firstly, we use word2vec to convert k-mers into number vectors to construct an input matrix. Secondly, we use convolutional neural network and bidirectional long short-term memory network to extract sequence features, and finally use the attention mechanism to extract relatively important features. In the task of predicting enhancers and their strengths, this method has improved to a certain extent in most evaluation indexes. In summary, we believe that this method provides new ideas in the analysis of enhancers.

Single cell RNA sequencing (scRNA-seq) enables researchers to characterize transcriptomic profiles at the single-cell resolution with increasingly high throughput. Clustering is a crucial step in single cell analysis. Clustering analysis of transcriptome profiled by scRNA-seq can reveal the heterogeneity and diversity of cells. However, single cell study still remains great challenges due to its high noise and dimension. Subspace clustering aims at discovering the intrinsic structure of data in unsupervised fashion. In this paper, we propose a deep sparse subspace clustering method scDSSC combining noise reduction and dimensionality reduction for scRNA-seq data, which simultaneously learns feature representation and clustering via explicit modelling of scRNA-seq data generation. Experiments on a variety of scRNA-seq datasets from thousands to tens of thousands of cells have shown that scDSSC can significantly improve clustering performance and facilitate the interpretability of clustering and downstream analysis. Compared to some popular scRNA-deq analysis methods, scDSSC outperformed state-of-the-art methods under various clustering performance metrics.

Extracting genes involved in cancer lesions from gene expression data is critical for cancer research and drug development. The method of feature selection has attracted much attention in the field of bioinformatics. Principal Component Analysis (PCA) is a widely used method for learning low-dimensional representation. Some variants of PCA have been proposed to improve the robustness and sparsity of the algorithm. However, the existing methods ignore the high-order relationships between data. In this paper, a new model named Robust Principal Component Analysis via Hypergraph Regularization (HRPCA) is proposed. In detail, HRPCA utilizes L2,1-norm to reduce the effect of outliers and make data sufficiently row-sparse. And the hypergraph regularization is introduced to consider the complex relationship among data. Important information hidden in the data are mined, and this method ensures the accuracy of the resulting data relationship information. Extensive experiments on multi-view biological data demonstrate that the feasible and effective of the proposed approach.

Numerous studies have shown that microRNAs are associated with the occurrence and development of human diseases. Thus, studying disease-associated miRNAs is significantly valuable to the prevention, diagnosis and treatment of diseases. In this paper, we proposed a novel method based on matrix completion and non-negative matrix factorization (MCNMF)for predicting disease-associated miRNAs. Due to the information inadequacy on miRNA similarities and disease similarities, we calculated the latter via two models, and introduced the Gaussian interaction profile kernel similarity. In addition, the matrix completion (MC)was employed to further replenish the miRNA and disease similarities to improve the prediction performance. And to reduce the sparsity of miRNA-disease association matrix, the method of weighted K nearest neighbor (WKNKN)was used, which is a pre-processing step. We also utilized non-negative matrix factorization (NMF)using dual${{\boldsymbol{L}}_{2,1}}$-norm, graph Laplacian regularization, and Tikhonov regularization to effectively avoid the overfitting during the prediction. Finally, several experiments and a case study were implemented to evaluate the effectiveness and performance of the proposed MCNMF model. The results indicated that our method could reliably and effectively predict disease-associated miRNAs.

MicroRNAs (miRNAs) are a class of non-coding RNAs that play critical role in many biological processes, such as cell growth, development, differentiation and aging. Increasing studies have revealed that miRNAs are closely involved in many human diseases. Therefore, the prediction of miRNA-disease associations is of great significance to the study of the pathogenesis, diagnosis and intervention of human disease. However, biological experimentally methods are usually expensive in time and money, while computational methods can provide an efficient way to infer the underlying disease-related miRNAs. In this study, we propose a novel method to predict potential miRNA-disease associations, called SVAEMDA. Our method mainly consider the miRNA-disease association prediction as semi-supervised learning problem. SVAEMDA integrates disease semantic similarity, miRNA functional similarity and respective Gaussian interaction profile (GIP) similarities. The integrated similarities are used to learn the representations of diseases and miRNAs. SVAEMDA trains a variational autoencoder based predictor by using known miRNA-disease associations, with the form of concatenated dense vectors. Reconstruction probability of the predictor is used to measure the correlation of the miRNA-disease pairs. Experimental results show that SVAEMDA outperforms other stat-of-the-art methods. AUC values of SVAEMDA of global leave-one-out cross validation (LOOCV) and 5-fold cross validation (5-fold CV) are 0.9464 and 0.9428 respectively. In addition, case studies of three common human diseases indicate that SVAEMDA obtains 100 percent of the top 50 predicted candidates in the benchmark databases. Therefore, SVAEMDA can efficiently and accurately predict the potential associations between diseases and miRNAs.

Identifying functional modules in protein-protein interaction (PPI) networks elucidates cellular organization and mechanism. Various methods have been proposed to identify the functional modules in PPI networks, but most of these methods do not consider the noisy links in PPI networks. They achieve a competitive performance on the PPI networks without noisy links, but the performance of these methods considerably deteriorates in the noisy PPI networks. Furthermore, the noisy links are inevitable in the PPI networks. In this paper, we propose a novel link-driven label propagation algorithm (LLPA) to identify functional modules in PPI networks. The LLPA first find link clusters in PPI networks, and then the functional modules are identified from the link clusters. Two strategies aimed to ensure the robustness of LLPA are proposed. One strategy involves the proposed LLPA updating the link labels in accordance with the designed weight of the link, which can reduce the incidence of noisy links. The other strategy involves the filtration of some noisy labels from the link clusters to further reduce the influence of noisy links. The performance evaluation on three real PPI networks shows that LLPA outperforms other eight state-of-the-art detection algorithms in terms of accuracy and robustness.

Recently, as a growing number of associations between microRNAs (miRNAs) and diseases are discovered, researchers gradually realize that miRNAs are closely related to several complicated biological processes and human diseases. Hence, it is especially important to construct availably models to infer associations between miRNAs and diseases. In this study, we presented Improved Graph Regression for miRNA-Disease Association Prediction (IGRMDA) to observe potential relationship between miRNAs and diseases. In order to reduce the inherent noise existing in the acquired biological datasets, we utilized matrix decomposition algorithm to process miRNA functional similarity and disease semantic similarity and then combining them with existing similarity information to obtain final miRNA similarity data and disease similarity data. Then, we applied miRNA-disease association data, miRNA similarity data and disease similarity data to form corresponding latent spaces. Furthermore, we performed improved graph regression algorithm in latent spaces, which included miRNA-disease association space, miRNA similarity space and disease similarity space. Non-negative matrix factorization and partial least squares were used in the graph regression process to obtain important related attributes. The cross validation experiments and case studies were also implemented to prove the effectiveness of IGRMDA, which showed that IGRMDA could predict potential associations between miRNAs and diseases.

The rapid development of single-cell RNA sequencing (scRNA-seq)technology reveals the gene expression status and gene structure of individual cells, reflecting the heterogeneity and diversity of cells. The traditional methods of scRNA-seq data analysis treat data as the same subspace, and hide structural information in other subspaces. In this paper, we propose a low-rank subspace ensemble clustering framework (LRSEC)to analyze scRNA-seq data. Assuming that the scRNA-seq data exist in multiple subspaces, the low-rank model is used to find the lowest rank representation of the data in the subspace. It is worth noting that the penalty factor of the low-rank kernel function is uncertain, and different penalty factors correspond to different low-rank structures. Moreover, the single cluster model is difficult to find the cellular structure of all datasets. To strengthen the correlation between model solutions, we construct a new ensemble clustering framework LRSEC by using the low-rank model as the basic learner. The LRSEC framework captures the global structure of data through low-rank subspaces, which has better clustering performance than a single clustering model. We validate the performance of the LRSEC framework on seven small datasets and one large dataset and obtain satisfactory results.

Identifying cell types is one of the main goals of single-cell RNA sequencing (scRNA-seq) analysis, and clustering is a common method for this item. However, the massive amount of data and the excess noise level bring challenge for single cell clustering. To address this challenge, in this paper, we introduced a novel method named single-cell clustering based on denoising autoencoder and graph convolution network (scCDG), which consists of two core models. The first model is a denoising autoencoder (DAE) used to fit the data distribution for data denoising. The second model is a graph autoencoder using graph convolution network (GCN), which projects the data into a low-dimensional space (compressed) preserving topological structure information and feature information in scRNA-seq data simultaneously. Extensive analysis on seven real scRNA-seq datasets demonstrate that scCDG outperforms state-of-the-art methods in some research sub-fields, including single cell clustering, visualization of transcriptome landscape, and trajectory inference.

Lots of experimental studies have revealed the significant associations between lncRNAs and diseases. Identifying accurate associations will provide a new perspective for disease therapy. Calculation-based methods have been developed to solve these problems, but these methods have some limitations. In this paper, we proposed an accurate method, named MLGCNET, to discover potential lncRNA-disease associations. Firstly, we reconstructed similarity networks for both lncRNAs and diseases using top k similar information, and constructed a lncRNA-disease heterogeneous network (LDN). Then, we applied Multi-Layer Graph Convolutional Network on LDN to obtain latent feature representations of nodes. Finally, the Extra Trees was used to calculate the probability of association between disease and lncRNA. The results of extensive 5-fold cross-validation experiments show that MLGCNET has superior prediction performance compared to the state-of-the-art methods. Case studies confirm the performance of our model on specific diseases. All the experiment results prove the effectiveness and practicality of MLGCNET in predicting potential lncRNA-disease associations.

In recent years, the association between synonymous mutations (SMs) and human diseases has been uncovered in many studies. It is a challenge for identifying deleterious SMs in the field of medical genomics. Although there are several computational methods proposed in the past years, the precise prediction of deleterious SMs is still challenging. In this work, we proposed a predictor named as EnDSM, which is an accurate method based on the ensemble framework. We explored multimodal features across four groups including functional score, conservation, splicing, and sequence features, and we then trained eight conceptually different machine learning classifiers for each of them, resulting in 32 base classification models. We further selected four base models referring to their prediction performance and the predictive probabilities of these base classification models were subsequently used as the input feature vectors of logistic regression classifier to construct the ensemble learning model. The results suggested that EnDSM achieved better performance comparing with other state-of-the-art predictors on the training and independent test datasets. We anticipate that our ensemble predictor EnDSM will become a valuable tool for deleterious SM prediction.The EnDSM server interface along with the benchmarking data sets are freely available athttp://bioinfo.ahu.edu.cn/EnDSM.

A growing number of clinical studies have provided substantial evidence of a close relationship between the microbe and the disease. Thus, it is necessary to infer potential microbe-disease associations. But traditional approaches use experiments to validate these associations that often spend a lot of materials and time. Hence, more reliable computational methods are expected to be applied to predict disease-associated microbes. In this article, an innovative mean for predicting microbe-disease associations is proposed, which is based on network consistency projection and label propagation (NCPLP). Given that most existing algorithms use the Gaussian interaction profile (GIP) kernel similarity as the similarity criterion between microbe pairs and disease pairs, in this model, Medical Subject Headings descriptors are considered to calculate disease semantic similarity. In addition, 16S rRNA gene sequences are borrowed for the calculation of microbe functional similarity. In view of the gene-based sequence information, we use two conventional methods (BLAST+ and MEGA7) to assess the similarity between each pair of microbes from different perspectives. Especially, network consistency projection is added to obtain network projection scores from the microbe space and the disease space. Ultimately, label propagation is utilized to reliably predict microbes related to diseases. NCPLP achieves better performance in various evaluation indicators and discovers a greater number of potential associations between microbes and diseases. Also, case studies further confirm the reliable prediction performance of NCPLP. To conclude, our algorithm NCPLP has the ability to discover these underlying microbe-disease associations and can provide help for biological study.

The exploration of single cell RNA-sequencing (scRNA-seq) technology generates a new perspective to analyze biological problems. One of the major applications of scRNA-seq data is to discover subtypes of cells by cell clustering. Nevertheless, it is challengeable for traditional methods to handle scRNA-seq data with high level of technical noise and notorious dropouts. To better analyze single cell data, a novel scRNA-seq data analysis model called Maximum correntropy criterion based Non-negative and Low Rank Representation (MccNLRR) is introduced. Specifically, the maximum correntropy criterion, as an effective loss function, is more robust to the high noise and large outliers existed in the data. Moreover, the low rank representation is proven to be a powerful tool for capturing the global and local structures of data. Therefore, some important information, such as the similarity of cells in the subspace, is also extracted by it. Then, an iterative algorithm on the basis of the half-quadratic optimization and alternating direction method is developed to settle the complex optimization problem. Before the experiment, we also analyze the convergence and robustness of MccNLRR. At last, the results of cell clustering, visualization analysis, and gene markers selection on scRNA-seq data reveal that MccNLRR method can distinguish cell subtypes accurately and robustly.

The development of single-cell RNA sequencing (scRNA-seq) technology has made it possible to measure gene expression levels at the resolution of a single cell, which further reveals the complex growth processes of cells such as mutation and differentiation. Recognizing cell heterogeneity is one of the most critical tasks in scRNA-seq research. To solve it, we propose a non-negative matrix factorization framework based on multi-subspace cell similarity learning for unsupervised scRNA-seq data analysis (MscNMF). MscNMF includes three parts: data decomposition, similarity learning, and similarity fusion. The three work together to complete the data similarity learning task. MscNMF can learn the gene features and cell features of different subspaces, and the correlation and heterogeneity between cells will be more prominent in multi-subspaces. The redundant information and noise in each low-dimensional feature space are eliminated, and its gene weight information can be further analyzed to calculate the optimal number of subpopulations. The final cell similarity learning will be more satisfactory due to the fusion of cell similarity information in different subspaces. The advantage of MscNMF is that it can calculate the number of cell types and the rank of Non-negative matrix factorization (NMF) reasonably. Experiments on eight real scRNA-seq datasets show that MscNMF can effectively perform clustering tasks and extract useful genetic markers. To verify its clustering performance, the framework is compared with other latest clustering algorithms and satisfactory results are obtained. The code of MscNMF is free available for academic (https://github.com/wangchuanyuan1/project-MscNMF).

Single-cell clustering is a crucial task of scRNA-seq analysis, which reveals the natural grouping of cells. However, due to the high noise and high dimension in scRNA-seq data, how to effectively and accurately identify cell types from a great quantity of cell mixtures is still a challenge. Considering this, in this paper, we propose a novel subspace clustering algorithm termed SLRRSC. This method is developed based on the low-rank representation model, and it aims to capture the global and local properties inherent in data. In order to make the LRR matrix describe the spatial relationship of samples more accurately, we introduce the manifold-based graph regularization and similarity constraint into the LRR-based method SLRRSC. The graph regularization can preserve the local geometric structure of the data in low-rank decomposition, so that the low-rank representation matrix contains more local structure information. By imposing similarity constraint on the low-rank matrix, the similarity information between sample pairs is further introduced into the SLRRSC model to improve the learning ability of low-rank method for global structure. At the same time, the similarity constraint makes the low-rank representation matrix symmetric, which makes it better interpretable in clustering application. We compare the effectiveness of the SLRRSC algorithm with other single-cell clustering methods on simulated data and real single-cell datasets. The results show that this method can obtain more accurate sample similarity matrix and effectively solve the problem of cell type recognition.

Non-negative Matrix Factorization (NMF) as a powerful dimension reduction tool, which is widely used in the bioinformatics field. However, the loss function of conventional NMF is sensitive to non-Gaussian noise and outliers. In addition, NMF-based algorithm overlooks the geometric structure of high dimensional data. To improve the robustness of NMF, we propose a novel method called Sparse Hyper-graph regularized Non-negative Matrix Factorization by Maximizing Correntropy (SHNMF-MCC) in this paper. Specifically, the maximum correntropy criterion replaces the Euclidean distance in the loss term of SHNMF-MCC, which can filter out the noise with large outliers. Moreover, the high-order geometric structure in more sample points is completely preserved in the low-dimensional manifold through the hyper-graph regularization. Meanwhile, the sparse constraint is applied to the loss function to reduce matrix complexity and analysis difficulty. Then, the complex optimization problem can be solved by a half-quadratic (HQ) optimization approach. Before carrying out experiments, we analyze the convergence of SHNMF-MCC. Sample clustering experiments on The Cancer Genome Atlas (TCGA) data and single cell RNA-sequencing (scRNA-seq) data verify that the proposed method is more robust and effective than other similar robust approaches.

Although synonymous mutations do not alter the encoded amino acids, they may impact protein function by interfering with the regulation of RNA splicing or altering transcript splicing. New progress on next-generation sequencing technologies has put the exploration of synonymous mutations at the forefront of precision medicine. Several approaches have been proposed for predicting the deleterious synonymous mutations specifically, but their performance is limited by imbalance of the positive and negative samples. In this study, we firstly expanded the number of samples greatly from various data sources and compared six undersampling strategies to solve the problem of the imbalanced datasets. The results suggested that cluster centroid is the most effective scheme. Secondly, we presented a computational model, undersampling scheme based method for deleterious synonymous mutation (usDSM) prediction, using 14-dimensional biology features and random forest classifier to detect the deleterious synonymous mutation. The results on the test datasets indicated that the proposed usDSM model can attain superior performance in comparison with other state-of-the-art machine learning methods. Lastly, we found that the deep learning model did not play a substantial role in deleterious synonymous mutation prediction through a lot of experiments, although it achieves superior results in other fields. In conclusion, we hope our work will contribute to the future development of computational methods for a more accurate prediction of the deleterious effect of human synonymous mutation. The web server of usDSM is freely accessible at http://usdsm.xialab.info/.

When this paper was originally published online, the lower part of Figure 2 was missing, and the web server name in Data availability Section was listed incorrectly. In addition, four values in Table 3 should have been shown in bold. The paper has been corrected online.

Predicting disease-related long non-coding RNAs (lncRNAs) is beneficial to finding of new biomarkers for prevention, diagnosis and treatment of complex human diseases. In this paper, we proposed a machine learning techniques-based classification approach to identify disease-related lncRNAs by graph auto-encoder (GAE) and random forest (RF) (GAERF). First, we combined the relationship of lncRNA, miRNA and disease into a heterogeneous network. Then, low-dimensional representation vectors of nodes were learned from the network by GAE, which reduce the dimension and heterogeneity of biological data. Taking these feature vectors as input, we trained a RF classifier to predict new lncRNA-disease associations (LDAs). Related experiment results show that the proposed method for the representation of lncRNA-disease characterizes them accurately. GAERF achieves superior performance owing to the ensemble learning method, outperforming other methods significantly. Moreover, case studies further demonstrated that GAERF is an effective method to predict LDAs.

MOTIVATION: MicroRNAs (miRNAs) are a class of non-coding RNAs that play critical roles in various biological processes. Many studies have shown that miRNAs are closely related to the occurrence, development and diagnosis of human diseases. Traditional biological experiments are costly and time consuming. As a result, effective computational models have become increasingly popular for predicting associations between miRNAs and diseases, which could effectively boost human disease diagnosis and prevention. RESULTS: We propose a novel computational framework, called AEMDA, to identify associations between miRNAs and diseases. AEMDA applies a learning-based method to extract dense and high-dimensional representations of diseases and miRNAs from integrated disease semantic similarity, miRNA functional similarity and heterogeneous related interaction data. In addition, AEMDA adopts a deep autoencoder that does not need negative samples to retrieve the underlying associations between miRNAs and diseases. Furthermore, the reconstruction error is used as a measurement to predict disease-associated miRNAs. Our experimental results indicate that AEMDA can effectively predict disease-related miRNAs and outperforms state-of-the-art methods. AVAILABILITY AND IMPLEMENTATION: The source code and data are available at https://github.com/CunmeiJi/AEMDA. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: As one of the deadliest diseases in the world, cancer is driven by a few somatic mutations that disrupt the normal growth of cells, and leads to abnormal proliferation and tumor development. The vast majority of somatic mutations did not affect the occurrence and development of cancer; thus, identifying the mutations responsible for tumor occurrence and development is one of the main targets of current cancer treatments. RESULTS: To effectively identify driver genes, we adopted a semi-local centrality measure and gene mutation effect function to assess the effect of gene mutations on changes in gene expression patterns. Firstly, we calculated the mutation score for each gene. Secondly, we identified differentially expressed genes (DEGs) in the cohort by comparing the expression profiles of tumor samples and normal samples, and then constructed a local network for each mutation gene using DEGs and mutant genes according to the protein-protein interaction network. Finally, we calculated the score of each mutant gene according to the objective function. The top-ranking mutant genes were selected as driver genes. We name the proposed method as mutations effect and network centrality. CONCLUSIONS: Four types of cancer data in The Cancer Genome Atlas were tested. The experimental data proved that our method was superior to the existing network-centric method, as it was able to quickly and easily identify driver genes and rare driver factors.

BACKGROUND: In recent years, various sequencing techniques have been used to collect biomedical omics datasets. It is usually possible to obtain multiple types of omics data from a single patient sample. Clustering of omics data plays an indispensable role in biological and medical research, and it is helpful to reveal data structures from multiple collections. Nevertheless, clustering of omics data consists of many challenges. The primary challenges in omics data analysis come from high dimension of data and small size of sample. Therefore, it is difficult to find a suitable integration method for structural analysis of multiple datasets. RESULTS: In this paper, a multi-view clustering based on Stiefel manifold method (MCSM) is proposed. The MCSM method comprises three core steps. Firstly, we established a binary optimization model for the simultaneous clustering problem. Secondly, we solved the optimization problem by linear search algorithm based on Stiefel manifold. Finally, we integrated the clustering results obtained from three omics by using k-nearest neighbor method. We applied this approach to four cancer datasets on TCGA. The result shows that our method is superior to several state-of-art methods, which depends on the hypothesis that the underlying omics cluster class is the same. CONCLUSION: Particularly, our approach has better performance than compared approaches when the underlying clusters are inconsistent. For patients with different subtypes, both consistent and differential clusters can be identified at the same time.

BACKGROUND: Circular RNAs (circRNAs) are a class of single-stranded RNA molecules with a closed-loop structure. A growing body of research has shown that circRNAs are closely related to the development of diseases. Because biological experiments to verify circRNA-disease associations are time-consuming and wasteful of resources, it is necessary to propose a reliable computational method to predict the potential candidate circRNA-disease associations for biological experiments to make them more efficient. RESULTS: In this paper, we propose a double matrix completion method (DMCCDA) for predicting potential circRNA-disease associations. First, we constructed a similarity matrix of circRNA and disease according to circRNA sequence information and semantic disease information. We also built a Gauss interaction profile similarity matrix for circRNA and disease based on experimentally verified circRNA-disease associations. Then, the corresponding circRNA sequence similarity and semantic similarity of disease are used to update the association matrix from the perspective of circRNA and disease, respectively, by matrix multiplication. Finally, from the perspective of circRNA and disease, matrix completion is used to update the matrix block, which is formed by splicing the association matrix obtained in the previous step with the corresponding Gaussian similarity matrix. Compared with other approaches, the model of DMCCDA has a relatively good result in leave-one-out cross-validation and five-fold cross-validation. Additionally, the results of the case studies illustrate the effectiveness of the DMCCDA model. CONCLUSION: The results show that our method works well for recommending the potential circRNAs for a disease for biological experiments.

miRNAs belong to small non-coding RNAs that are related to a number of complicated biological processes. Considerable studies have suggested that miRNAs are closely associated with many human diseases. In this study, we proposed a computational model based on Similarity Constrained Matrix Factorization for miRNA-Disease Association Prediction (SCMFMDA). In order to effectively combine different disease and miRNA similarity data, we applied similarity network fusion algorithm to obtain integrated disease similarity (composed of disease functional similarity, disease semantic similarity and disease Gaussian interaction profile kernel similarity) and integrated miRNA similarity (composed of miRNA functional similarity, miRNA sequence similarity and miRNA Gaussian interaction profile kernel similarity). In addition, the L2 regularization terms and similarity constraint terms were added to traditional Nonnegative Matrix Factorization algorithm to predict disease-related miRNAs. SCMFMDA achieved AUCs of 0.9675 and 0.9447 based on global Leave-one-out cross validation and five-fold cross validation, respectively. Furthermore, the case studies on two common human diseases were also implemented to demonstrate the prediction accuracy of SCMFMDA. The out of top 50 predicted miRNAs confirmed by experimental reports that indicated SCMFMDA was effective for prediction of relationship between miRNAs and diseases.

microRNAs (miRNAs) are small non-coding RNAs related to a number of complicated biological processes. A growing body of studies have suggested that miRNAs are closely associated with many human diseases. It is meaningful to consider disease-related miRNAs as potential biomarkers, which could greatly contribute to understanding the mechanisms of complex diseases and benefit the prevention, detection, diagnosis and treatment of extraordinary diseases. In this study, we presented a novel model named Graph Convolutional Autoencoder for miRNA-Disease Association Prediction (GCAEMDA). In the proposed model, we utilized miRNA-miRNA similarities, disease-disease similarities and verified miRNA-disease associations to construct a heterogeneous network, which is applied to learn the embeddings of miRNAs and diseases. In addition, we separately constructed miRNA-based and disease-based sub-networks. Combining the embeddings of miRNAs and diseases, graph convolutional autoencoder (GCAE) was utilized to calculate association scores of miRNA-disease on two sub-networks, respectively. Furthermore, we obtained final prediction scores between miRNAs and diseases by adopting an average ensemble way to integrate the prediction scores from two types of subnetworks. To indicate the accuracy of GCAEMDA, we applied different cross validation methods to evaluate our model whose performances were better than the state-of-the-art models. Case studies on a common human diseases were also implemented to prove the effectiveness of GCAEMDA. The results demonstrated that GCAEMDA was beneficial to infer potential associations of miRNA-disease.

Protein-protein interactions play essential roles in various biological progresses. Identifying protein interaction sites can facilitate researchers to understand life activities and therefore will be helpful for drug design. However, the number of experimental determined protein interaction sites is far less than that of protein sites in protein-protein interaction or protein complexes. Therefore, the negative and positive samples are usually imbalanced, which is common but bring result bias on the prediction of protein interaction sites by computational approaches. In this work, we presented three imbalance data processing strategies to reconstruct the original dataset, and then extracted protein features from the evolutionary conservation of amino acids to build a predictor for identification of protein interaction sites. On a dataset with 10,430 surface residues but only 2,299 interface residues, the imbalance dataset processing strategies can obviously reduce the prediction bias, and therefore improve the prediction performance of protein interaction sites. The experimental results show that our prediction models can achieve a better prediction performance, such as a prediction accuracy of 0.758, or a high F-measure of 0.737, which demonstrated the effectiveness of our method.

Pathogenicity-related studies are of great importance in understanding the pathogenesis of complex diseases and improving the level of clinical medicine. This work proposed a bioinformatics scheme to analyze cancer-related gene mutations, and try to figure out potential genes associated with diseases from the protein domain-domain interaction network. Herein, five measures of the principle of centrality lethality had been adopted to implement potential correlation analysis, and prioritize the significance of genes. This method was further applied to KEGG pathway analysis by taking the malignant melanoma as an example. The experimental results show that 25 domains can be found, and 18 of them have high potential to be pathogenically important related to malignant melanoma. Finally, a web-based tool, named Human Cancer Related Domain Interaction Network Analyzer, is developed for potential pathogenic genes prioritization for 26 types of human cancers, and the analysis results can be visualized and downloaded online.

Non-negative matrix factorization (NMF) has become one of the most powerful methods for clustering and feature selection. However, the performance of the traditional NMF method severely degrades when the data contain noises and outliers or the manifold structure of the data is not taken into account. In this article, a novel method called correntropy-based hypergraph regularized NMF (CHNMF) is proposed to solve the above problem. Specifically, we use the correntropy instead of the Euclidean norm in the loss term of CHNMF, which will improve the robustness of the algorithm. And the hypergraph regularization term is also applied to the objective function, which can explore the high-order geometric information in more sample points. Then, the half-quadratic (HQ) optimization technique is adopted to solve the complex optimization problem of CHNMF. Finally, extensive experimental results on multi-cancer integrated data indicate that the proposed CHNMF method is superior to other state-of-the-art methods for clustering and feature selection.

The rapid development of single cell RNA sequencing (scRNA-seq) has made it possible to study the association between cells and genes at molecular resolution. When the follow-up analysis is carried out, it is often difficult to extract the cell information in high-dimensional space because of the high gene dimension in single-cell sequencing, which leads to inaccurate results in the follow-up analysis. To solve the problem, we propose a method called locally manifold non-negative matrix factorization based on centroid for scRNA-seq data analysis (MNMFC). MNMFC is a similarity modeling scheme based on locally manifold, which can map cell association in high dimensional space. Through similarity learning based on locally manifold and non-negative matrix decomposition (NMF) algorithm, the data in high-dimensional space can be mapped to low-dimensional space, which provides help for downstream clustering analysis. The performance of the model was validated experimentally on 10 scRNA-seq datasets. Compared with other nine advanced single-cell clustering methods, whether it is a comprehensive analysis or an individual analysis of the dataset, MNMFC has achieved encouraging results.

Automatic breast cancer grading methods based on HE stained pathological images can be summarized into two categories. The first category is to use learning-based methods to directly extract the features of the pathological image for breast cancer grading. However, unlike the coarse-grained problem of breast cancer classification, grading of breast Invasive Ductal Carcinoma (IDC) is a fine-grained classification problem. Only using general methods cannot classify IDC well. The second category is to conduct the three evaluation criteria of Nottingham Grading System (NGS) separately, and then integrate the results of the three criteria to obtain the final IDC grading result. However, NGS is only a semi-quantitative evaluation method. The inherent medical motivation of NGS is to grade IDC with the help of nuclei-related features. In this paper, we proposed a nuclei-aware network for IDC grading in pathological images. The entire network achieves an effect similar to the attention mechanism in end-to-end learning, so as to learn fine-grained and nuclei-related feature representations for IDC grading. It should to be pointed out that our method can emphasize custom areas, thus providing a way to model medical knowledge into the network structure. This is different from the general attention mechanism that cannot artificially control the area of attention. Experimental results show that the performance of proposed method is better than the state-of-the-art.

Synonymous mutations do not change the encoded amino acids but may alter the structure or function of an mRNA in ways that impact gene function. Advances in next generation sequencing technologies have detected numerous synonymous mutations in the human genome. Several computational models have been proposed to predict deleterious synonymous mutations, which have greatly facilitated the development of this important field. Consequently, there is an urgent need to assess the state-of-the-art computational methods for deleterious synonymous mutation prediction to further advance the existing methodologies and to improve performance. In this regard, we systematically compared a total of 10 computational methods (including specific method for deleterious synonymous mutation and general method for single nucleotide mutation) in terms of the algorithms used, calculated features, performance evaluation and software usability. In addition, we constructed two carefully curated independent test datasets and accordingly assessed the robustness and scalability of these different computational methods for the identification of deleterious synonymous mutations. In an effort to improve predictive performance, we established an ensemble model, named Prediction of Deleterious Synonymous Mutation (PrDSM), which averages the ratings generated by the three most accurate predictors. Our benchmark tests demonstrated that the ensemble model PrDSM outperformed the reviewed tools for the prediction of deleterious synonymous mutations. Using the ensemble model, we developed an accessible online predictor, PrDSM, available at http://bioinfo.ahu.edu.cn:8080/PrDSM/. We hope that this comprehensive survey and the proposed strategy for building more accurate models can serve as a useful guide for inspiring future developments of computational methods for deleterious synonymous mutation prediction.

DNA-binding hot spot residues of proteins are dominant and fundamental interface residues that contribute most of the binding free energy of protein-DNA interfaces. As experimental methods for identifying hot spots are expensive and time consuming, computational approaches are urgently required in predicting hot spots on a large scale. In this work, we systematically assessed a wide variety of 114 features from a combination of the protein sequence, structure, network and solvent accessible information and their combinations along with various feature selection strategies for hot spot prediction. We then trained and compared four commonly used machine learning models, namely, support vector machine (SVM), random forest, Naïve Bayes and k-nearest neighbor, for the identification of hot spots using 10-fold cross-validation and the independent test set. Our results show that (1) features based on the solvent accessible surface area have significant effect on hot spot prediction; (2) different but complementary features generally enhance the prediction performance; and (3) SVM outperforms other machine learning methods on both training and independent test sets. In an effort to improve predictive performance, we developed a feature-based method, namely, PrPDH (Prediction of Protein-DNA binding Hot spots), for the prediction of hot spots in protein-DNA binding interfaces using SVM based on the selected 10 optimal features. Comparative results on benchmark data sets indicate that our predictor is able to achieve generally better performance in predicting hot spots compared to the state-of-the-art predictors. A user-friendly web server for PrPDH is well established and is freely available at http://bioinfo.ahu.edu.cn:8080/PrPDH.

Abstract Background The aberrant expression of microRNAs is closely connected to the occurrence and development of a great deal of human diseases. To study human diseases, numerous effective computational models that are valuable and meaningful have been presented by researchers. Results Here, we present a computational framework based on graph Laplacian regularizedL2,1-nonnegative matrix factorization (GRL2,1-NMF) for inferring possible human disease-connected miRNAs. First, manually validated disease-connected microRNAs were integrated, and microRNA functional similarity information along with two kinds of disease semantic similarities were calculated. Next, we measured Gaussian interaction profile (GIP) kernel similarities for both diseases and microRNAs. Then, we adopted a preprocessing step, namely, weighted K nearest known neighbours (WKNKN), to decrease the sparsity of the miRNA-disease association matrix network. Finally, theGRL2,1-NMF framework was used to predict links between microRNAs and diseases. Conclusions The new method (GRL2, 1-NMF) achieved AUC values of 0.9280 and 0.9276 in global leave-one-out cross validation (global LOOCV) and five-fold cross validation (5-CV), respectively, showing that GRL2, 1-NMF can powerfully discover potential disease-related miRNAs, even if there is no known associated disease.

Abstract Background As a machine learning method with high performance and excellent generalization ability, extreme learning machine (ELM) is gaining popularity in various studies. Various ELM-based methods for different fields have been proposed. However, the robustness to noise and outliers is always the main problem affecting the performance of ELM. Results In this paper, an integrated method named correntropy induced loss based sparse robust graph regularized extreme learning machine (CSRGELM) is proposed. The introduction of correntropy induced loss improves the robustness of ELM and weakens the negative effects of noise and outliers. By using the L2,1-norm to constrain the output weight matrix, we tend to obtain a sparse output weight matrix to construct a simpler single hidden layer feedforward neural network model. By introducing the graph regularization to preserve the local structural information of the data, the classification performance of the new method is further improved. Besides, we design an iterative optimization method based on the idea of half quadratic optimization to solve the non-convex problem of CSRGELM. Conclusions The classification results on the benchmark dataset show that CSRGELM can obtain better classification results compared with other methods. More importantly, we also apply the new method to the classification problems of cancer samples and get a good classification effect.

Epistasis learning, which is aimed at detecting associations between multiple Single Nucleotide Polymorphisms (SNPs) and complex diseases, has gained increasing attention in genome wide association studies. Although much work has been done on mapping the SNPs underlying complex diseases, there is still difficulty in detecting epistatic interactions due to the lack of heuristic information to expedite the search process. In this study, a method EACO is proposed to detect epistatic interactions based on the ant colony optimization (ACO) algorithm, the highlights of which are the introduced heuristic information, fitness function, and a candidate solutions filtration strategy. The heuristic information multi-SURF* is introduced into EACO for identifying epistasis, which is incorporated into ant-decision rules to guide the search with linear time. Two functionally complementary fitness functions, mutual information and the Gini index, are combined to effectively evaluate the associations between SNP combinations and the phenotype. Furthermore, a strategy for candidate solutions filtration is provided to adaptively retain all optimal solutions which yields a more accurate way for epistasis searching. Experiments of EACO, as well as three ACO based methods (AntEpiSeeker, MACOED, and epiACO) and four commonly used methods (BOOST, SNPRuler, TEAM, and epiMODE) are performed on both simulation data sets and a real data set of age-related macular degeneration. Results indicate that EACO is promising in identifying epistasis.

In recent years, with the diversity and variability of cancer information, the multi-omics data have been applied in various fields. Many existing models of principal component analysis can only process single data, which makes limitations on cancer research. Therefore, in this paper, a new model called integrative principal component analysis (IPCA) is proposed to achieve the unification of multi-omics data. In addition, in order to preserve the high-order manifold structure between the data, an integrative hypergraph regularization principal component analysis (IHPCA) is further proposed by applying the hypergraph regularization constraint. The effectiveness of IHPCA method is tested on four multi-omics datasets. Experimental results show that the proposed method has better performance than other representative methods on sample clustering and common expression genes (co-expression genes) network analysis.

BACKGROUND: Predicting miRNA-disease associations (MDAs) is time-consuming and expensive. It is imminent to improve the accuracy of prediction results. So it is crucial to develop a novel computing technology to predict new MDAs. Although some existing methods can effectively predict novel MDAs, there are still some shortcomings. Especially when the disease matrix is processed, its sparsity is an important factor affecting the final results. RESULTS: -norm are introduced to our method to achieve the highest AUC value than other advanced methods. CONCLUSIONS: 5-fold cross validation is used to evaluate our method, and simulation experiments are used to predict novel associations on Gold Standard Dataset. Finally, our prediction accuracy is better than other existing advanced methods. Therefore, our approach is effective and feasible in predicting novel MDAs.

BACKGROUND: Predicting meaningful miRNA-disease associations (MDAs) is costly. Therefore, an increasing number of researchers are beginning to focus on methods to predict potential MDAs. Thus, prediction methods with improved accuracy are under development. An efficient computational method is proposed to be crucial for predicting novel MDAs. For improved experimental productivity, large biological datasets are used by researchers. Although there are many effective and feasible methods to predict potential MDAs, the possibility remains that these methods are flawed. RESULTS: A simple and effective method, known as Nearest Profile-based Collaborative Matrix Factorization (NPCMF), is proposed to identify novel MDAs. The nearest profile is introduced to our method to achieve the highest AUC value compared with other advanced methods. For some miRNAs and diseases without any association, we use the nearest neighbour information to complete the prediction. CONCLUSIONS: To evaluate the performance of our method, five-fold cross-validation is used to calculate the AUC value. At the same time, three disease cases, gastric neoplasms, rectal neoplasms and colonic neoplasms, are used to predict novel MDAs on a gold-standard dataset. We predict the vast majority of known MDAs and some novel MDAs. Finally, the prediction accuracy of our method is determined to be better than that of other existing methods. Thus, the proposed prediction model can obtain reliable experimental results.

BACKGROUND: The recognition of protein interaction sites is of great significance in many biological processes, signaling pathways and drug designs. However, most sites on protein sequences cannot be defined as interface or non-interface sites because only a small part of protein interactions had been identified, which will cause the lack of prediction accuracy and generalization ability of predictors in protein interaction sites prediction. Therefore, it is necessary to effectively improve prediction performance of protein interaction sites using large amounts of unlabeled data together with small amounts of labeled data and background knowledge today. RESULTS: In this work, three semi-supervised support vector machine-based methods are proposed to improve the performance in the protein interaction sites prediction, in which the information of unlabeled protein sites can be involved. Herein, five features related with the evolutionary conservation of amino acids are extracted from HSSP database and Consurf Sever, i.e., residue spatial sequence spectrum, residue sequence information entropy and relative entropy, residue sequence conserved weight and residual Base evolution rate, to represent the residues within the protein sequence. Then three predictors are built for identifying the interface residues from protein surface using three types of semi-supervised support vector machine algorithms. CONCLUSION: The experimental results demonstrated that the semi-supervised approaches can effectively improve prediction performance of protein interaction sites when unlabeled information is involved into the predictors and one of them can achieve the best prediction performance, i.e., the accuracy of 70.7%, the sensitivity of 62.67% and the specificity of 78.72%, respectively. With comparison to the existing studies, the semi-supervised models show the improvement of the predication performance.

Tumor samples clustering based on biomolecular data is a hot issue of cancer classifications discovery. How to extract the valuable information from high dimensional genomic data is becoming an urgent problem in tumor samples clustering. In this paper, we introduce manifold regularization into low-rank representation model and present a novel method named Mixed-norm Laplacian regularized Low-Rank Representation (MLLRR) to identify the differentially expressed genes for tumor clustering based on gene expression data. Then, in order to advance the accuracy and stability of tumor clustering, we establish the clustering model based on Penalized Matrix Decomposition (PMD) and propose a novel cluster method named MLLRR-PMD. In this method, the cancer clustering research includes three steps. First, the matrix of gene expression data is decomposed into a low rank representation matrix and a sparse matrix by MLLRR. Second, the differentially expressed genes are identified based on the sparse matrix. Finally, the PMD is applied to cluster the samples based on the differentially expressed genes. The experiment results on simulation data and real genomic data illustrate that MLLRR method enhances the robustness to outliers and achieves remarkable performance in the extraction of differentially expressed genes.

Principal component analysis (PCA) has been used to study the pathogenesis of diseases. To enhance the interpretability of classical PCA, various improved PCA methods have been proposed to date. Among these, a typical method is the so-called sparse PCA, which focuses on seeking sparse loadings. However, the performance of these methods is still far from satisfactory due to their limitation of using unsupervised learning methods; moreover, the class ambiguity within the sample is high. To overcome this problem, this paper developed a new PCA method, which is named the supervised discriminative sparse PCA (SDSPCA). The main innovation of this method is the incorporation of discriminative information and sparsity into the PCA model. Specifically, in contrast to the traditional sparse PCA, which imposes sparsity on the loadings, here, sparse components are obtained to represent the data. Furthermore, via the linear transformation, the sparse components approximate the given label information. On the one hand, sparse components improve interpretability over the traditional PCA, while on the other hand, they are have discriminative abilities suitable for classification purposes. A simple algorithm is developed, and its convergence proof is provided. SDSPCA has been applied to the common-characteristic gene selection and tumor classification on multiview biological data. The sparsity and classification performance of SDSPCA are empirically verified via abundant, reasonable, and effective experiments, and the obtained results demonstrate that SDSPCA outperforms other state-of-the-art methods.

Non-negative Matrix Factorization (NMF), a classical method for dimensionality reduction, has been applied in many fields. It is based on the idea that negative numbers are physically meaningless in various data-processing tasks. Apart from its contribution to conventional data analysis, the recent overwhelming interest in NMF is due to its newly discovered ability to solve challenging data mining and machine learning problems, especially in relation to gene expression data. This survey paper mainly focuses on research examining the application of NMF to identify differentially expressed genes and to cluster samples, and the main NMF models, properties, principles, and algorithms with its various generalizations, extensions, and modifications are summarized. The experimental results demonstrate the performance of the various NMF algorithms in identifying differentially expressed genes and clustering samples.

With the advent of sequencing technology, numerous gene expression data are generated. Identifying differentially expressed genes play an important role in the gene therapy of cancer patients. As an useful mathematical tool, nonnegative matrix factorization (NMF) has been successfully used for identifying differentially expressed genes. In this paper, a novel method named robust graph regularized sparse orthogonal nonnegative matrix factorization (RGSON) is proposed and used for identifying differentially expressed genes, which introduces manifold learning, L1and orthogonal constraints into the objective function. In particular, L2,1-norm minimization is enforced on the objective function to improve the robustness of the algorithm. To prove the validity of the algorithm, experiments on the real genomic dataset are conducted. The results show that RGSON performs more effective than many other methods for identifying differentially expressed genes.

Low-rank representation (LRR) via rank minimization is a high efficiency method for capturing low-dimensional structure embedded in high-dimensional data. However, minimizing the rank of a matrix is NP-hard. In this paper, robust truncated nuclear norm low-rank representation regularized by L2,1-norm method (RTLRR) is proposed. The truncated nuclear norm is introduced to replace the nuclear norm to approximate the rank function. At the same time, L2,1-norm is used to regularize the sparse matrix to achieve better sparse effect of the algorithm. The proposed method is divided into two steps. Firstly, we do singular value decomposition (SVD) to the original data matrix. Then we apply the truncated nuclear norm and L2,1-norm constraints to subproblems and use inexact augmented Lagrange multiplier method to solve subproblems. Finally, the genes with high scores will be identified as differentially expressed genes according to the sparse matrix. The results on The Cancer Genome Atlas (TCGA) data illustrate that the effectiveness of RTLRR method outperforms many methods.

Non-negative Matrix Factorization (NMF) is widely used as a data dimensionality reduction tool. However, the assumption of most conventional NMF-based methods is that the gene expression data are only destroyed by Gaussian noise. In practice, the gene expression data are unavoidably destroyed by sparse noise. Although Sparsity-Regularized Robust NMF by using L1/2constraint (L1/2-RNMF) can achieve satisfactory results when the sparse noise exists, it does not consider the intrinsic geometric structure in data. Hence, we introduce graph regularization into L1/2-RNMF. In this paper, we developed a novel NMF method named Graph regularized Robust Nonnegative Matrix Factorization (GrRNMF), which mainly consists of two aspects: Firstly, the Gaussian noise and sparse noise are modeled, respectively. Secondly, it can reveal the geometric information in data by adding graph regularization term. Extensive experimental results on The Cancer Genome Atlas (TCGA) data indicate that the GrRNMF method has higher accuracy than other state-of-the-art methods in samples clustering and the selection of differentially expressed genes.

One major goal of large-scale cancer omics study is to understand molecular mechanisms of cancer and find new biomedical targets. To deal with the high-dimensional multidimensional cancer omics data (DNA methylation, mRNA expression, etc.), which can be used to discover new insight on identifying cancer subtypes, clustering methods are usually used to find an effective low-dimensional subspace of the original data and then cluster cancer samples in the reduced subspace. However, due to data-type diversity and big data volume, few methods can integrate these data and map them into an effective low-dimensional subspace. In this paper, we develop a dimension-reduction and data-integration method for indentifying cancer subtypes, named Scluster. First, Scluster, respectively, projects the different original data into the principal subspaces by an adaptive sparse reduced-rank regression method. Then, a fused patient-by-patient network is obtained for these subgroups through a scaled exponential similarity kernel method. Finally, candidate cancer subtypes are identified using spectral clustering method. We demonstrate the efficiency of our Scluster method using three cancers by jointly analyzing mRNA expression, miRNA expression, and DNA methylation data. The evaluation results and analyses show that Scluster is effective for predicting survival and identifies novel cancer subtypes of large-scale multi-omics data.

Recent research shows that characteristic gene selection based on gene expression data remains faced with considerable challenges. This is primarily because vast amount of gene expression data have been generated with the development of gene detection technology. Nonetheless, the recognition rate and reliability of gene selection still need to be improved. In this paper, we propose a novel constrained method: robust graph regularized discriminative nonnegative matrix factorization (RGDNMF) for characteristic gene selection. The method mainly includes two aspects: firstly, we incorporate both intrinsic geometrical structure and discriminative label information into the NMF model. Secondly, we adopt L2,1 -norm minimization to both the error function and the regularization term which is robust to noises and outliers in gene data. Furthermore we present the multiplicative update rules and the convergence proof. Our experiments demonstrate that RGDNMF is far more effective than other existing methods.

Tumor clustering based on biomolecular data plays a very important role for cancer classifications discovery. To further improve the robustness, stability and accuracy of tumor clustering, we develop a novel dimension reduction method named p-norm singular value decomposition (PSVD) to seek a low-rank approximation matrix to the bimolecular data. To enhance the robustness to outliers, the Lp-norm is taken as the error function and the Schatten p-norm is used as the regularization function in our optimization model. To evaluate the performance of PSVD, Kmeans clustering method is then employed for tumor clustering based on the low-rank approximation matrix. The extensive experiments are performed on gene expression dataset and cancer genome dataset respectively. All experimental results demonstrate that the PSVD-based method outperforms many existing methods. Especially it is experimentally proved that the proposed method is efficient for processing higher dimensional data with good robustness and superior time performance.

Recently, feature extraction and dimensionality reduction have become fundamental tools for many data mining tasks, especially for processing high-dimensional data such as genome data. In this paper, a new feature extraction method based on sparse singular value decomposition (SSVD) is developed. SSVD algorithm is applied to extract differentially expressed genes from two different genome datasets that are all from The Cancer Genome Atlas (TCGA), and then the extracted genes are evaluated by the tools based on Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis. As a gene extraction method, SSVD is also compared with some existing feature extraction methods such as independent component analysis, the p-norm robust feature extraction and sparse principal component analysis. The experimental GO analysis results show that SSVD method outperforms the competitive algorithms. The KEGG analysis results demonstrate the genes which participate in the pathways in cancer. The elaborate experiments prove that SSVD is an effective feature selection method compared with the competitive methods. The KEGG analysis results may provide a meaningful reference to carry out further study for professionals in the field of biomedical science.

Robust Principal Component Analysis (RPCA) is an efficient method in the selection of differentially expressed genes. However, nuclear norm minimizes all singular values simultaneously, so it may not be the best solution to replace the low-rank function. In this paper, the truncated nuclear norm is introduced. And a new method named Truncated nuclear norm regularized Robust Principal Component Analysis (TRPCA) is proposed. The method decomposes the observation matrix of genomic data into a low-rank matrix and a sparse matrix. The differentially expressed genes can be selected according to the sparse matrix. The experimental results on the The Cancer Genome Atlas (TCGA) data illustrate that the TRPCA method outperforms other state-of-the-art methods in the selection of differentially expressed genes.

Pathway-based drug discovery overcomes the disadvantages of the “one drug-one target” method, which aims to find the effective drugs to act on single targets. The current method “iPaD” identities the drug-pathway association pairs by taking the lasso-type penalty on the drug-pathway association matrix. In order to enhance the robustness of the methods and be more effective to find the novel drug-pathway association pairs, we introduce a new method named “L2,1-iPaD”. Compared with the iPaD method, we impose the L2,1-norm constraint on the drug-pathway association coefficient matrix. By applying our method to a real widely datasets (CCLE dataset), we demonstrate that our method is superior to the iPaD method. And our method can obtain the smaller P-values than the iPaD method by performing permutation test to assess the significance of the identified drug-pathway association pairs. More importantly, compared with the iPaD method, our method can identify larger numbers of validated drug-pathway association pairs. The experimental results on the real dataset demonstrate the effectiveness of our method.

In this paper, we presented a network-based method, named DriverFinder, by filtering frequently mutated genes just because of their large size, and comparing tumor expression with normal expression data to obtain gene expression outliers which are more likely to be cancer genes. Then greedy algorithm was applied to prioritize candidate driver genes. The proposed method can not only indentify frequently mutated genes, but also novel and infrequently mutated driver genes.

BACKGROUND: Detecting and visualizing nonlinear interaction effects of single nucleotide polymorphisms (SNPs) or epistatic interactions are important topics in bioinformatics since they play an important role in unraveling the mystery of "missing heritability". However, related studies are almost limited to pairwise epistatic interactions due to their methodological and computational challenges. RESULTS: We develop CINOEDV (Co-Information based N-Order Epistasis Detector and Visualizer) for the detection and visualization of epistatic interactions of their orders from 1 to n (n ≥ 2). CINOEDV is composed of two stages, namely, detecting stage and visualizing stage. In detecting stage, co-information based measures are employed to quantify association effects of n-order SNP combinations to the phenotype, and two types of search strategies are introduced to identify n-order epistatic interactions: an exhaustive search and a particle swarm optimization based search. In visualizing stage, all detected n-order epistatic interactions are used to construct a hypergraph, where a real vertex represents the main effect of a SNP and a virtual vertex denotes the interaction effect of an n-order epistatic interaction. By deeply analyzing the constructed hypergraph, some hidden clues for better understanding the underlying genetic architecture of complex diseases could be revealed. CONCLUSIONS: Experiments of CINOEDV and its comparison with existing state-of-the-art methods are performed on both simulation data sets and a real data set of age-related macular degeneration. Results demonstrate that CINOEDV is promising in detecting and visualizing n-order epistatic interactions. CINOEDV is implemented in R and is freely available from R CRAN: http://cran.r-project.org and https://sourceforge.net/projects/cinoedv/files/ .

BACKGROUND: Cancer is a complex disease which is characterized by the accumulation of genetic alterations during the patient's lifetime. With the development of the next-generation sequencing technology, multiple omics data, such as cancer genomic, epigenomic and transcriptomic data etc., can be measured from each individual. Correspondingly, one of the key challenges is to pinpoint functional driver mutations or pathways, which contributes to tumorigenesis, from millions of functional neutral passenger mutations. RESULTS: In this paper, in order to identify driver genes effectively, we applied a generalized additive model to mutation profiles to filter genes with long length and constructed a new gene-gene interaction network. Then we integrated the mutation data and expression data into the gene-gene interaction network. Lastly, greedy algorithm was used to prioritize candidate driver genes from the integrated data. We named the proposed method Length-Net-Driver (LNDriver). CONCLUSIONS: Experiments on three TCGA datasets, i.e., head and neck squamous cell carcinoma, kidney renal clear cell carcinoma and thyroid carcinoma, demonstrated that the proposed method was effective. Also, it can identify not only frequently mutated drivers, but also rare candidate driver genes.

With the development of deep sequencing technologies, many RNA-Seq data have been generated. Researchers have proposed many methods based on the sparse theory to identify the differentially expressed genes from these data. In order to improve the performance of sparse principal component analysis, in this paper, we propose a novel class-information-based sparse component analysis (CISCA) method which introduces the class information via a total scatter matrix. First, CISCA normalizes the RNA-Seq data by using a Poisson model to obtain their differential sections. Second, the total scatter matrix is gotten by combining the between-class and within-class scatter matrices. Third, we decompose the total scatter matrix by using singular value decomposition and construct a new data matrix by using singular values and left singular vectors. Then, aiming at obtaining sparse components, CISCA decomposes the constructed data matrix by solving an optimization problem with sparse constraints on loading vectors. Finally, the differentially expressed genes are identified by using the sparse loading vectors. The results on simulation and real RNA-Seq data demonstrate that our method is effective and suitable for analyzing these data.

Many methods have been considered for gene selection and analysis of gene expression data. Nonetheless, there still exists the considerable space for improving the explicitness and reliability of gene selection. To this end, this paper proposes a novel method named robust graph regularized non-negative matrix factorization for characteristic gene selection using gene expression data, which mainly contains two aspects: Firstly, enforcing L21-norm minimization on error function which is robust to outliers and noises in data points. Secondly, it considers that the samples lie in low-dimensional manifold which embeds in a high-dimensional ambient space, and reveals the data geometric structure embedded in the original data. To demonstrate the validity of the proposed method, we apply it to gene expression data sets involving various human normal and tumor tissue samples and the results demonstrate that the method is effective and feasible.

MOTIVATION: Tremendous amount of omics data being accumulated poses a pressing challenge of meta-analyzing the heterogeneous data for mining new biological knowledge. Most existing methods deal with each gene independently, thus often resulting in high false positive rates in detecting differentially expressed genes (DEG). To our knowledge, no or little effort has been devoted to methods that consider dependence structures underlying transcriptomics data for DEG identification in meta-analysis context. RESULTS: This article proposes a new meta-analysis method for identification of DEGs based on joint non-negative matrix factorization (jNMFMA). We mathematically extend non-negative matrix factorization (NMF) to a joint version (jNMF), which is used to simultaneously decompose multiple transcriptomics data matrices into one common submatrix plus multiple individual submatrices. By the jNMF, the dependence structures underlying transcriptomics data can be interrogated and utilized, while the high-dimensional transcriptomics data are mapped into a low-dimensional space spanned by metagenes that represent hidden biological signals. jNMFMA finally identifies DEGs as genes that are associated with differentially expressed metagenes. The ability of extracting dependence structures makes jNMFMA more efficient and robust to identify DEGs in meta-analysis context. Furthermore, jNMFMA is also flexible to identify DEGs that are consistent among various types of omics data, e.g. gene expression and DNA methylation. Experimental results on both simulation data and real-world cancer data demonstrate the effectiveness of jNMFMA and its superior performance over other popular approaches. AVAILABILITY AND IMPLEMENTATION: R code for jNMFMA is available for non-commercial use via http://micblab.iim.ac.cn/Download/. CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Microarray techniques have been used to delineate cancer groups or to identify candidate genes for cancer prognosis. As such problems can be viewed as classification ones, various classification methods have been applied to analyze or interpret gene expression data. In this paper, we propose a novel method based on robust principal component analysis (RPCA) to classify tumor samples of gene expression data. Firstly, RPCA is utilized to highlight the characteristic genes associated with a special biological process. Then, RPCA and RPCA+LDA (robust principal component analysis and linear discriminant analysis) are used to identify the features. Finally, support vector machine (SVM) is applied to classify the tumor samples of gene expression data based on the identified features. Experiments on seven data sets demonstrate that our methods are effective and feasible for tumor classification.

Cancer prognosis is an important clinical practice in cancer medicine and is an important factor in developing personalized medicine. But till now, researches focus on developing recurrence risk indices that tell poor or good survival for given cancer patients. These indices, however, are insufficient and elusive in the clinic. In this paper, we propose to predict survival time of cancer patients using pattern recognition approach, which is more informative and favorable to clinicians and patients in clinical practice. We conduct an extensive survey of pattern recognition methods for the prognosis based on real-world benchmark microarray data sets. In particular, various types of data preprocessing methods and various types of classification models are introduced and examined for predicting survival time of lung cancer based on gene expression. The experimental results show that pattern recognition method can provide a feasible and efficient way to predict survival time of cancer patients. It is expected that the pattern classification-based strategy opens a new paradigm of cancer prognosis for predicting survival time of cancer patients in the clinic.

BACKGROUND: Identifying protein-protein interactions (PPIs) is essential for elucidating protein functions and understanding the molecular mechanisms inside the cell. However, the experimental methods for detecting PPIs are both time-consuming and expensive. Therefore, computational prediction of protein interactions are becoming increasingly popular, which can provide an inexpensive way of predicting the most likely set of interactions at the entire proteome scale, and can be used to complement experimental approaches. Although much progress has already been achieved in this direction, the problem is still far from being solved and new approaches are still required to overcome the limitations of the current prediction models. RESULTS: In this work, a sequence-based approach is developed by combining a novel Multi-scale Continuous and Discontinuous (MCD) feature representation and Support Vector Machine (SVM). The MCD representation gives adequate consideration to the interactions between sequentially distant but spatially close amino acid residues, thus it can sufficiently capture multiple overlapping continuous and discontinuous binding patterns within a protein sequence. An effective feature selection method mRMR was employed to construct an optimized and more discriminative feature set by excluding redundant features. Finally, a prediction model is trained and tested based on SVM algorithm to predict the interaction probability of protein pairs. CONCLUSIONS: When performed on the yeast PPIs data set, the proposed approach achieved 91.36% prediction accuracy with 91.94% precision at the sensitivity of 90.67%. Extensive experiments are conducted to compare our method with the existing sequence-based method. Experimental results show that the performance of our predictor is better than several other state-of-the-art predictors, whose average prediction accuracy is 84.91%, sensitivity is 83.24%, and precision is 86.12%. Achieved results show that the proposed approach is very promising for predicting PPI, so it can be a useful supplementary tool for future proteomics studies. The source code and the datasets are freely available at http://csse.szu.edu.cn/staff/youzh/MCDPPI.zip for academic use.

Differential coexpression analysis usually requires the definition of 'distance' or 'similarity' between measured datasets. Until now, the most common choice is Pearson correlation coefficient. However, Pearson correlation coefficient is sensitive to outliers. Biweight midcorrelation is considered to be a good alternative to Pearson correlation since it is more robust to outliers. In this paper, we introduce to use Biweight Midcorrelation to measure 'similarity' between gene expression profiles, and provide a new approach for gene differential coexpression analysis. Firstly, we calculate the biweight midcorrelation coefficients between all gene pairs. Then, we filter out non-informative correlation pairs using the 'half-thresholding' strategy and calculate the differential coexpression value of gene, The experimental results on simulated data show that the new approach performed better than three previously published differential coexpression analysis (DCEA) methods. Moreover, we use the maximum clique analysis to gene subset included genes identified by our approach and previously reported T2D-related genes, many additional discoveries can be found through our method.

Adverse side effects of drug-drug interactions induced by human cytochrome P450 (CYP) inhibition play crucial roles in drug discovery. It is urgent and challenging to develop computational methods to efficiently and accurately predict the inhibitive effect of a compound against a specific CYP isoform. In this work we present a novel EELM (ensemble of extreme learning machine) model to predict CYP inhibition. Particularly, extreme learning machine (ELM) and fingerprint descriptors are firstly used to build the weak learning machines. And then EELM is constructed by combining the outputs of each individual ELM using majority voting strategy. Experimental results demonstrate that the proposed method yields good results compared with the existing methods.

More and more studies have shown many complex diseases are contributed jointly by alterations of numerous genes. In this paper, we propose a gene differential coexpression analysis algorithm in the level of gene sets and apply the algorithm to a publicly available type 2 diabetes (T2D) expression dataset. The experimental results on simulated data show that the new approach performed well. Moreover, we apply the new approach to clinical data, many additional discoveries can be found through our method.

How to identify a set of genes that are relevant to a key biological process is an important issue in current molecular biology. In this paper, we propose a novel method to discover differentially expressed genes based on robust principal component analysis (RPCA). In our method, we treat the differentially and non-differentially expressed genes as perturbation signals S and low-rank matrix A, respectively. Perturbation signals S can be recovered from the gene expression data by using RPCA. To discover the differentially expressed genes associated with special biological progresses or functions, the scheme is given as follows. Firstly, the matrix D of expression data is decomposed into two adding matrices A and S by using RPCA. Secondly, the differentially expressed genes are identified based on matrix S. Finally, the differentially expressed genes are evaluated by the tools based on Gene Ontology. A larger number of experiments on hypothetical and real gene expression data are also provided and the experimental results show that our method is efficient and effective.

Eigengene extracted by independent component analysis (ICA) is one kind of effective feature for tumor classification. In this letter, a novel tumor classification approach is proposed by using eigengene and support vector machine (SVM) based classifier committee learning (CCL) algorithm. In this method, a strategy of random feature subspace division is designed to improve the diversity of weaker classifiers. Gene expression data constructed by different feature subspaces are modeled by ICA, respectively. And the corresponding eigengene sets extracted by the ICA algorithm are used as the inputs of the weaker SVM classifiers. Moreover, a strategy of Bayesian sum rule (BSR) is designed to integrate the outputs of the weaker SVM classifiers, and used to provide a final decision for the tumor category. Experimental results on three DNA microarray datasets demonstrate that the proposed method is effective and feasible for tumor classification.

A reliable and accurate identification of the type of tumors is crucial to the proper treatment of cancers. In recent years, it has been shown that sparse representation (SR) by l1-norm minimization is robust to noise, outliers and even incomplete measurements, and SR has been successfully used for classification. This paper presents a new SR-based method for tumor classification using gene expression data. A set of metasamples are extracted from the training samples, and then an input testing sample is represented as the linear combination of these metasamples by l1-regularized least square method. Classification is achieved by using a discriminating function defined on the representation coefficients. Since l1-norm minimization leads to a sparse solution, the proposed method is called metasample-based SR classification (MSRC). Extensive experiments on publicly available gene expression data sets show that MSRC is efficient for tumor classification, achieving higher accuracy than many existing representative schemes.

A reliable and precise identification of the type of tumors is crucial to the effective treatment of cancer. With the rapid development of microarray technologies, tumor clustering based on gene expression data is becoming a powerful approach to cancer class discovery. In this paper, we apply the penalized matrix decomposition (PMD) to gene expression data to extract metasamples for clustering. The extracted metasamples capture the inherent structures of samples belong to the same class. At the same time, the PMD factors of a sample over the metasamples can be used as its class indicator in return. Compared with the conventional methods such as hierarchical clustering (HC), self-organizing maps (SOM), affinity propagation (AP) and nonnegative matrix factorization (NMF), the proposed method can identify the samples with complex classes. Moreover, the factor of PMD can be used as an index to determine the cluster number. The proposed method provides a reasonable explanation of the inconsistent classifications made by the conventional methods. In addition, it is able to discover the modules in gene expression data of conterminous developmental stages. Experiments on two representative problems show that the proposed PMD-based method is very promising to discover biological phenotypes.

With the advent of DNA microarrays, it is now possible to use the microarrays data for tumor classification. Yet previous works have not use the nonnegative information of gene expression data. In this paper, we propose a new method for tumor classification using gene expression data. In this method, we first select genes using nonnegative matrix factorization (NMF) and sparse NMF (SNMF). Then we extract features of the selected gene data by virtue of NMF and SNMF. At last, support vector machines (SVM) was applied to classify the tumor samples based on the extracted features. To better fit for classification aim, a modified SNMF algorithm is also proposed. The experimental results on three microarray datasets show that the method is efficient and feasible.

Tumor clustering is becoming a powerful method in cancer class discovery. Nonnegative matrix factorization (NMF) has shown advantages over other conventional clustering techniques. Nonetheless, there is still considerable room for improving the performance of NMF. To this end, in this paper, gene selection and explicitly enforcing sparseness are introduced into the factorization process. Particularly, independent component analysis is employed to select a subset of genes so that the effect of irrelevant or noisy genes can be reduced. The NMF and its extensions, sparse NMF and NMF with sparseness constraint, are then used for tumor clustering on the selected genes. A series of elaborate experiments are performed by varying the number of clusters and the number of selected genes to evaluate the cooperation between different gene selection settings and NMF-based clustering. Finally, the experiments on three representative gene expression datasets demonstrated that the proposed scheme can achieve better clustering results.

In this letter, a standard postnonlinear blind source separation algorithm is proposed, based on the MISEP method, which is widely used in linear and nonlinear independent component analysis. To best suit a wide class of postnonlinear mixtures, we adapt the MISEP method to incorporate a priori information of the mixtures. In particular, a group of three-layered perceptrons and a linear network are used as the unmixing system to separate sources in the postnonlinear mixtures, and another group of three-layered perceptron is used as the auxiliary network. The learning algorithm for the unmixing system is then obtained by maximizing the output entropy of the auxiliary network. The proposed method is applied to postnonlinear blind source separation of both simulation signals and real speech signals, and the experimental results demonstrate its effectiveness and efficiency in comparison with existing methods.

MOTIVATION: Microarrays are capable of determining the expression levels of thousands of genes simultaneously. One important application of gene expression data is classification of samples into categories. In combination with classification methods, this technology can be useful to support clinical management decisions for individual patients, e.g. in oncology. Standard statistic methodologies in classification or prediction do not work well when the number of variables p (genes) far too exceeds the number of samples n. So, modification of existing statistical methodologies or development of new methodologies is needed for the analysis of microarray data. RESULTS: This paper proposes a new method for tumor classification using gene expression data. In this method, we first employ independent component analysis to model the gene expression data, then apply optimal scoring algorithm to classify them. Further speaking, this approach can first make full use of the high-order statistical information contained in the gene expression data. Second, this approach also employs regularized regression models to handle the situation of large numbers of correlated predictor variables. Finally, the predictive models are developed for classifying tumors based on the entire gene expression profile. To show the validity of the proposed method, we apply it to classify four DNA microarray datasets involving various human normal and tumor tissue samples. The experimental results show that the method is efficient and feasible. AVAILABILITY: Matlab scripts are available on request.

This paper proposes a new method for tumor classification using gene expression data, which mainly contains three steps. Firstly, the original DNA microarray gene expression data are selected using t-statistics. Secondly, the selected genes are modeled by Independent Component Analysis (ICA). Finally, Support Vector Machine (SVM) is used to classify the modeling data. To show the validity of the proposed method, we apply it to classify two DNA microarray data sets involving various human normal and tumor tissue samples. The experimental results show that the method is efficient and feasible.

This paper proposes an extended non-negative sparse coding (NNSC) neural network method for image compression. This method can exploit the NNSC algorithm to obtain transform-based compression schemes adapted to standard natural image classes, which results from the statistical properties of natural image data. In particular, several methods of image compression such as linear principal component analysis (PCA), wavelet-based analysis, independent component analysis (ICA), etc., are evaluated and compared based on both the standard signal to noise ratio (SNR) and picture quality scale (PQS) criteria. The simulation results show that, in the case of using a fixed block by block scanning a natural image randomly, the quality of a compressed image obtained by our extended NNSC compression algorithm indeed outperforms the one obtained by other algorithms mentioned above.

In this paper, a nonlinear blind source separation system with post-nonlinear mixing; model, and an unsupervised learning algorithm for the parameters of this separating system are presented for blind inversion of Wiener system for single source. The proposed method firstly changes the deconvolution part of Wiener system into a special case of linear blind source separation (BSS). Then the nonlinear BSS system is applied to derive the source signal. The proposed nonlinear BSS method can dynamically estimate the nonlinearity of mixing model and adapt to the cumulative probability function (CPF) of sources. Finally, experimental results demonstrate that our proposed method is effective and efficient for the problems addressed.