IEEE transactions on computational biology and bioinformatics最新文献

{"title":"Partition-Aware Joint Sparse Precision Matrix Estimation for Cancer Diagnosis.","authors":"Rwan Ahmed, Kang Jiang, Weilai Chi, Fang-Xiang Wu","doi":"10.1109/TCBBIO.2026.3723674","DOIUrl":"10.1109/TCBBIO.2026.3723674","url":null,"abstract":"<p><p>Precision matrices, which encode gene regulatory networks, play an important role for cancer diagnosis from transcriptomics data. Cancer subtypes often form partially related families that share regulatory structure to varying degrees, yet most multi-condition network estimation methods treat subtype relatedness as either absent or uniform. This mistreatment is particularly limiting in high-dimensional transcriptomic studies, where subtype relationships are uncertain and sample sizes are small. We propose Partition-Aware Joint Sparse Precision Matrix Estimation (PA-JSPME), a unified framework that jointly estimates subtype-specific precision matrices while learning latent clusters of related subtypes directly from data. The method introduces a partition-aware fusion penalty that selectively encourages similarity within inferred clusters while allowing networks from different clusters to diverge. To reduce shrinkage bias and preserve strong regulatory signals, PA-JSPME employs the Smoothly Clipped Absolute Deviation (SCAD) regularization. Computational scalability is achieved through a quadratic surrogate likelihood and an efficient alternating optimization algorithm combining k-means partition updates with an ADMM-MM solver. Experiments on synthetical datasets demonstrate that PA-JSPME accurately recovers both latent subtype structure and class-specific networks across a range of dimensionality settings. Applications to pediatric and adult brain tumor transcriptomics datasets yield biologically coherent subtype groupings, sparse and interpretable regulatory networks, and improved cancer diagnosis performance compared to existing joint graphical modeling approaches. Furthermore experiments on two additional real-world transcriptomic datasets illustrate the generalizability of our proposed method.</p>","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148762852","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ScCLC: A Flexible Contrastive Learning Framework for Single-cell Multi-omics Data Clustering.","authors":"Zhenlan Liang, Ruiqing Zheng, Huayu Tao, Yuxuan Chen, Min Li","doi":"10.1109/TCBBIO.2026.3720975","DOIUrl":"10.1109/TCBBIO.2026.3720975","url":null,"abstract":"<p><p>The rapid development of single-cell joint profiling technologies enables the simultaneous measurement of multiple molecular modalities from the same cell, providing unprecedented opportunities to characterize cellular heterogeneity. However, effectively integrating heterogeneous and high-dimensional multi-omics data remains a fundamental challenge for accurate cell clustering. In this work, we propose scCLC, a topology-aware contrastive learning framework for clustering single-cell multi-omics data. scCLC adopts contrastive learning as the backbone for cell representation learning and introduces a dedicated multi-view data augmentation strategy to address modality-specific characteristics. By exploiting the intrinsic cell-cell topological structures constructed from multi-omics data, scCLC identifies informative positive pairs for self-supervised training, which encourages the learned representations to be more cluster-discriminative. Extensive experiments on multiple paired datasets demonstrate the effectiveness of scCLC for clustering single-cell multi-omics data. Visualization analyses further indicate that scCLC is capable of distinguishing rare cell populations in highly imbalanced datasets. Moreover, case studies on single-cell triple-omics datasets illustrate that scCLC can be readily extended to integrate additional modalities, underscoring its flexibility and scalability for multi-omics data analysis. The source code can be downloaded from https://github.com/CSUBioGroup/scCLC.</p>","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148687576","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"GPCR-SLM: Small Language Model-Based Classification of GPCRs Using Knowledge Distillation Technique.","authors":"Fairuz Shadmani Shishir","doi":"10.1109/TCBBIO.2026.3718339","DOIUrl":"10.1109/TCBBIO.2026.3718339","url":null,"abstract":"<p><p>Accurate protein family classification is essential since proteins within the same family share conserved structural domains and biochemical functions that deter mine their biological roles. G protein-coupled receptors (GPCRs) represent one of the largest and most diverse protein families in eukaryotes, serving as targets for ap proximately 35% of FDA-approved drugs. While traditional sequence alignment methods, such as BLAST, provide foundational tools for identifying homologous sequences, they exhibit limited accuracy in distinguishing closely related GPCR families with low sequence homology. Recently, deep learning approaches offer promising accuracy; however, they employ fixed-size classification architectures that force newly discovered protein families into pre-existing categories, preventing the recognition of novel families and limiting scalability as the protein universe expands. In this work, we present a scalable machine learning framework GPCR SLM, that classifies GPCRs across 86 distinct families using a lightweight transformer model optimized through knowledge distillation. Our approach achieved an overall ac curacy of 99%, significantly outperforming BLAST (86.1%) and HMMER (91%), while demonstrating substantial computational efficiency with an average speedup of 33.5× compared to large protein language models. These results demonstrate the effectiveness of combining distilled protein language models with flexible classification frameworks for high-resolution functional annotation.</p>","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148623867","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"TREPP: Tandem Repeat Expansion Pathogenicity Prediction via Stacked CatBoost and Context-Aware Sequence Features.","authors":"Minghua Xu, Kang Hu, Chao Deng, Peng Ni, Jianxin Wang","doi":"10.1109/TCBBIO.2026.3712297","DOIUrl":"10.1109/TCBBIO.2026.3712297","url":null,"abstract":"<p><p>Predicting the pathogenic potential of tandem repeat (TR) expansions is critical for understanding neurogenetic disorders but remains challenging due to data scarcity, extreme class imbalance, and complex sequence dependencies. Most existing approaches primarily rely on gene-centric signals, often overlooking intrinsic sequence properties that mechanistically modulate instability. This paper proposes TREPP, an interpretable ensemble learning framework designed to prioritize candidate high-risk TR loci. TREPP combines genomic location information with local sequence properties to represent the biological context of expansion. First, the framework extracts a comprehensive feature set including gene proximity, GC content differences, and local repeat density patterns. Then, to mitigate label scarcity, TREPP employs a multi-sampling strategy that trains multiple CatBoost base learners on diversified negative subsets. Finally, a sparse logistic regression meta-learner aggregates the calibrated outputs of these base models to produce a final prioritization score. Experimental results on a curated benchmark dataset show that TREPP achieves an AUPRC of 94.71%, exceeding the state-of-the-art method RExPRT by 6.11 percentage points on the fixed benchmark split. Furthermore, interpretability and bias-control analyses support the interpretation that TREPP leverages biologically plausible signals, making it a useful prioritization framework for identifying candidate high-risk TR loci. TREPP is publicly available at https://github.com/minghuaxu/TREPP.</p>","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148440783","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"scVAEAT: An Integrative Attention-Augmented Variational Autoencoder for Predicting Single-Cell Perturbation Responses.","authors":"Binhua Tang, Yujia Zhang, Yiyao Chen, Xinyu Gao, Mengyao Mao","doi":"10.1109/TCBBIO.2026.3710792","DOIUrl":"10.1109/TCBBIO.2026.3710792","url":null,"abstract":"<p><p>Understanding how individual cells respond to genetic or environmental perturbations is crucial for deciphering disease mechanisms and advancing precision medicine. However, predicting single-cell perturbation responses from standard single-cell RNA sequencing (scRNA-seq) data remains challenging because of cellular heterogeneity and the lack of paired pre- and post-perturbation samples in the dataset. Here, we present scVAEAT, a novel deep learning framework that integrates variational autoencoders with attention-enhanced optimal transport to accurately predict gene expression variations following perturbations, such as interferon-β stimulation or parasitic infection. scVAEAT introduces two key innovations: (1) an attention-augmented encoding module that captures multiscale feature representations through combined global and local attention mechanisms, and (2) an attention-based optimal transport (OTA) strategy that aligns unperturbed and perturbed cell states without requiring paired data. Evaluated on benchmark human datasets, including peripheral blood mononuclear cells (PBMC) and intestinal epithelial cells, scVAEAT consistently outperformed existing methods in predicting perturbation-induced expression changes, achieving mean R² scores of up to 0.97 across diverse cell types. Ablation studies confirmed the contribution of each module to model performance, whereas robustness analyses demonstrated stability across hyperparameter settings. Notably, scVAEAT accurately recapitulated known interferon-stimulated gene responses and infection-associated markers, underscoring its high biological fidelity. By in silico perturbation modeling and validation experiments using unpaired scRNA-seq data, scVAEAT offers a powerful tool for dissecting the regulatory mechanisms underlying human genetic and environmental responses, with broad implications for functional genomics and drug development.</p>","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148407914","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Improving Cancer Driver Gene Prediction using Biological knowledge-guided Prompts for LLM.","authors":"Tao Yuan, Wei Peng, Wei Dai, Xiaodong Fu, Li Liu, Lijun Liu, Shikang Deng, Wei Lan","doi":"10.1109/TCBBIO.2026.3710090","DOIUrl":"10.1109/TCBBIO.2026.3710090","url":null,"abstract":"<p><p>Accurately identifying cancer driver genes is crucial for understanding cancer and developing therapies. However, challenges such as limited sample sizes and insufficient differentiation of gene characteristics hinder current approaches. Large language models (LLMs) possess a wealth of biomedical knowledge offering the potential to enhance gene representations. However, effectively integrating LLMs into cancer driver gene prediction requires careful prompt engineering and multiomics fusion strategies. This study introduces a novel framework, called Bioprompt, that leverages biological knowledge-guided prompts for LLM to improve cancer driver gene identification. We utilize Gene Ontology(GO) information to guide the design of prompts for LLMs, enabling them to generate relevant gene semantic representations. Semantic representations are then encoded through a text encoder and a variational autoencoder. Subsequently, we integrate these LLM-derived gene semantic features with existing model-generated gene features via contrastive learning. Finally, a logistic regression model is employed to combine these two feature types for the identification of cancer driver genes. Validation on pancancer and 15 individual cancer datasets shows improvement over existing models. Ablation studies confirm the critical role of GO-guided prompts in generating valuable gene semantic information. Independent test set tests have demonstrated the generalization ability of our method. Furthermore, by analyzing the gene functional vocabulary generated by the LLM, we gain valuable insights into cancer related gene functions and new perspectives on cancer mechanisms.</p>","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148383525","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Protein Complex Recognition Method Based on Dynamic Hypergraph Neural Network.","authors":"Peng Li, Zirui Chen, Juan Huang, Aijing Luo, Hui Min","doi":"10.1109/TCBBIO.2026.3698134","DOIUrl":"10.1109/TCBBIO.2026.3698134","url":null,"abstract":"<p><p>Accurate identification of protein complexes has significant scientific and practical value in disease research, bioengineering, and understanding life activities. Current identification algorithms are based on graph structures and do not consider the high-order nature of interactions between proteins, resulting in insufficient identification quality and efficiency. Considering the advantages of hypergraphs in handling high-order data and computational efficiency, a protein complex identification algorithm based on a dynamic hypergraph neural network (PCI-DHNN) is proposed. First, considering the dynamic and time-varying characteristics of protein networks, the protein network is sliced into time slices using the activity cycle of proteins. Then, dynamic protein subnetworks are constructed based on hypergraph theory. On this basis, a hypergraph convolutional operator is designed for the feature learning of protein subnetworks to obtain a membership matrix describing the potential relationships between proteins and complexes. Finally, the protein complexes were identified using a Bernoulli mixture model. Experimental simulation results on four protein network datasets of yeast species show that compared with current typical protein complex identification algorithms, the PCI-DHNN performs better in terms of recall, precision, F1-score, coverage rate, and functional enrichment, and has higher reliability.</p>","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":"1608-1621"},"PeriodicalIF":0.0,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148146048","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimized Gene Selection Using Nomadic People and Salp Swarm Algorithms for Cancer Detection.","authors":"Sadyaa Fahad Jabar, M A Balafar, Ali Jameel Hashim","doi":"10.1109/TCBBIO.2026.3687099","DOIUrl":"10.1109/TCBBIO.2026.3687099","url":null,"abstract":"<p><p>Accurately detecting cancer through gene expression analysis is crucial for early diagnosis and effective treatment. However, gene expression data's high dimensionality and redundancy pose significant challenges, such as overfitting and computational inefficiency. To address these issues, we propose a hybrid feature selection framework that integrates a filter-wrapper approach with swarm intelligence for optimized gene selection. The proposed method utilizes the Nomadic People Optimizer (NPO) in conjunction with Mutual Information (MI) to identify a relevant subset of genes from high-dimensional datasets. An optimized Support Vector Machine (SVM) is employed to further enhance classification accuracy, with its hyperparameters fine-tuned using an enhanced Salp Swarm Algorithm (SSA) incorporating a crossover operator. This hybrid approach not only reduces the search space but also mitigates overfitting by leveraging the exploration and exploitation capabilities of the NPO and SSA. Experimental results on five cancer gene expression datasets-lung adenocarcinoma (LUAD), breast cancer (GSE2034), glioblastoma multiforme (GBM), ovarian cancer (GSE2109), and colorectal adenocarcinoma (COAD)-demonstrate that the proposed NPO-SSVM achieves classification accuracies ranging from 91.25% to 97.02% with AUC-ROC values between 0.85 and 0.97. The framework achieves an average feature reduction of 51% while outperforming state-of-the-art methods (GA, PSO, GWO, SSA) by 3-12% in classification accuracy and 15% in computational efficiency. These findings confirm that NPO-SSVM provides a robust and efficient solution for gene selection in cancer detection, offering significant advancements for personalized medicine and early diagnosis.</p>","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":"1549-1560"},"PeriodicalIF":0.0,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147795791","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Context-Aware Self-Training Framework for Cell Type Annotation Using Marker Genes.","authors":"Hegang Chen, Yuyin Lu, Yanghui Rao","doi":"10.1109/TCBBIO.2026.3688205","DOIUrl":"10.1109/TCBBIO.2026.3688205","url":null,"abstract":"<p><p>Single-cell annotation is a fundamental task in the analysis of single-cell data, and one promising research direction relies on the marker gene information accumulated in biology. Recently, self-training strategies have been introduced into the field, which significantly improve the annotation accuracy by iteratively optimizing the model. However, existing methods have not yet systematically explored how to construct self-training frameworks that are more applicable to single-cell data. To this end, we propose a context-aware self-training model named CSSTA. Firstly, the contextual information of marker genes is introduced to enhance the compatibility of marker genes with different single-cell datasets to generate high-quality pseudo-labels. Secondly, high- and low-confidence pseudo-labels recognition and supervision strategies more applicable to single-cell data are designed that can better guide the optimization of the model. Finally, the insight of the single-cell foundation model on cell-cell association information is introduced. Experiments demonstrate that the introduction of marker gene contextual information significantly improves the ability to recognize cell-cell type associations with heuristic-based strategies. Benchmark experiments show that CSSTA significantly outperforms state-of-the-art baselines. Notably, we demonstrate the potential of CSSTA for hierarchical cellular annotation by extending it to hierarchies.</p>","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":"1515-1526"},"PeriodicalIF":0.0,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147795762","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"FGAIM: Identifying Drug-Target Activation and Inhibition Mechanisms via Inductive Graph Neural Networks Based on Fine-Grained Interaction Strategies.","authors":"Yi Tang, Yongxian Fan, Guicong Sun, Mengxin Zheng","doi":"10.1109/TCBBIO.2026.3695928","DOIUrl":"10.1109/TCBBIO.2026.3695928","url":null,"abstract":"<p><p>Distinguishing the activation and inhibition mechanisms between drugs and targets can reveal the potential regulatory pathways of target functions, which is crucial in drug discovery and development. Although numerous deep learning-based computational methods have been proposed, most of them extract drug and target features independently while neglecting interactions between drug molecules and protein residues. In addition, existing methods approach drugs in a relatively simple way and predominantly rely on protein sequence information. Deep learning, particularly graph neural networks (GNNs), has demonstrated unique advantages in processing data with complex graph-structured relationships. Motivated by these limitations, this study proposes a novel computational method named FGAIM, which designed to effectively identify activation and inhibition mechanisms between drugs and targets. First, a multi-scale GNN module in FGAIM is employed to learn expressive drug molecular embeddings. At the same time, protein representations are constructed by integrating pre-trained language model (PLM) embeddings and structural information derived from 3D conformations. Subsequently, FGAIM leverages a GraphSAGE module to extract features from both the primary drug graph and the fine-grained drug-protein interaction graph. Finally, the resulting drug and target embeddings are fused and fed into a Multilayer Perceptron (MLP) for classification prediction. Comparative experiments on two public datasets demonstrate that FGAIM significantly outperforms existing computational approaches and exhibits strong generalization capabilities. Further case studies reveal the advantages of FGAIM in mining previously unrecognized activation/inhibition relationships. Building upon the fine-grained interactions, we further explored the model's interpretability at the molecular structural level through analysis of attention weights. Additionally, visualization analyses confirm that FGAIM effectively captures underlying structural patterns in the data and identifies discriminative features across different categories.</p>","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":"1671-1683"},"PeriodicalIF":0.0,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147990809","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}