IEEE transactions on computational biology and bioinformatics最新文献

MMB-DTI: A Multimodal State Space Framework with Bidirectional Cross-Attention for Drug-Target Interaction Prediction. MMB-DTI：用于药物-靶标相互作用预测的双向交叉关注的多模态空间框架。

IEEE transactions on computational biology and bioinformatics Pub Date : 2026-10-26 DOI: 10.1109/TCBBIO.2026.3746102

Fang Fang, Jinjin Li, Chengye Huang, Rao Zeng, Leyi Wei

{"title":"MMB-DTI: A Multimodal State Space Framework with Bidirectional Cross-Attention for Drug-Target Interaction Prediction.","authors":"Fang Fang, Jinjin Li, Chengye Huang, Rao Zeng, Leyi Wei","doi":"10.1109/TCBBIO.2026.3746102","DOIUrl":"https://doi.org/10.1109/TCBBIO.2026.3746102","url":null,"abstract":"Accurate drug-target interaction (DTI) prediction is fundamental to computational drug discovery. However, existing deep learning methods typically oversimplify biochemical entities as 1D sequences, neglecting crucial spatial topologies, and incur prohibitive quadratic computational costs when modeling long-range protein dependencies. To overcome these bottlenecks, we propose MMB-DTI, a novel multimodal framework that synergistically fuses evolutionary and chemical semantics, extracted via pre-trained language models (ESM-2 and ChemBERTa), with structural topologies derived from graph neural networks. At its core, the framework introduces a hybrid Mamba-BiLSTM architecture, leveraging a selective state space model for linear-time encoding of ultra-long-range protein sequences while precisely preserving local sequential motifs. Furthermore, a bidirectional cross-attention module simulates the biological \"induced-fit\" mechanism, dynamically capturing reciprocal and fine-grained binding affinities. Comprehensive evaluations across six benchmark datasets (Davis, KIBA, DrugBank, Enzymes, Ion Channels, and GPCRs) demonstrate that MMB-DTI achieves highly competitive predictive performance, significantly outperforming state-of-the-art baselines. Our results highlight that coupling linear-time state space modeling with topological complementarity yields a robust and discriminative paradigm for deciphering complex drug-target associations.","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149281063","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}

引用次数: 0

MMPKAN-DTA: A Multimodal Framework for Drug-Target Affinity Prediction Using Pretrained Models and Kolmogorov-Arnold Networks. MMPKAN-DTA：使用预训练模型和Kolmogorov-Arnold网络进行药物-靶标亲和力预测的多模态框架。

IEEE transactions on computational biology and bioinformatics Pub Date : 2026-10-26 DOI: 10.1109/TCBBIO.2026.3746983

Muhammad Habibulla Alamin, Md Youshuf Khan Rakib, Md Sifat Ali, Qichang Zhao, Jianxin Wang

{"title":"MMPKAN-DTA: A Multimodal Framework for Drug-Target Affinity Prediction Using Pretrained Models and Kolmogorov-Arnold Networks.","authors":"Muhammad Habibulla Alamin, Md Youshuf Khan Rakib, Md Sifat Ali, Qichang Zhao, Jianxin Wang","doi":"10.1109/TCBBIO.2026.3746983","DOIUrl":"https://doi.org/10.1109/TCBBIO.2026.3746983","url":null,"abstract":"Drug-target affinity (DTA) prediction is crucial for advancing drug discovery and development, facilitating candidate prioritization and reducing experimental costs. However, accurate affinity prediction remains challenging, and most deep learning models often fail to generalize in scenarios involving novel drugs or targets (known as cold-start scenarios). This limitation is due to simplified feature representations, limited exploitation of complementary information, and insufficient capacity to model the complex nonlinear relationships in heterogeneous inputs. To overcome these limitations, we introduce a multimodal deep learning framework, MMPKAN-DTA, that combines pretrained features with GNN-derived features to learn representations for both drugs and proteins. The model incorporates a multimodal feature fusion block and Kolmogorov-Arnold Networks (KANs), which enhance representational expressivity by exploiting complementarity and nonlinear approximation capacity, thereby enabling a more effective and efficient model for drug-target affinity predictions. Comprehensive experiments on three benchmark datasets demonstrate that MMPKAN-DTA outperforms baseline methods and exhibits strong generalization ability. Moreover, a case study on the epidermal growth factor receptor (EGFR) and tyrosine-protein kinase ABL1 highlights the effectiveness of MMPKAN-DTA in predicting affinities for unseen drug-target pairs, supporting its applicability as a promising tool for drug discovery and development.","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149281137","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}

引用次数: 0

3.2-Approximation for Reversal Distance toward String Adjacency Recovery. 3.2字符串邻接恢复反转距离的逼近。

IEEE transactions on computational biology and bioinformatics Pub Date : 2026-10-23 DOI: 10.1109/TCBBIO.2026.3746274

Siqi Jiang, Tiantian Li, Daming Zhu

引用次数: 0

Hypergraph Masked Autoencoder for Drug Repositioning. 用于药物重定位的超图掩码自编码器。

IEEE transactions on computational biology and bioinformatics Pub Date : 2026-10-23 DOI: 10.1109/TCBBIO.2026.3746316

Meiyu Duan, Haijia Bi, Wenhui Xie, Ren Fu, Hai Cui, Yijia Zhang

{"title":"Hypergraph Masked Autoencoder for Drug Repositioning.","authors":"Meiyu Duan, Haijia Bi, Wenhui Xie, Ren Fu, Hai Cui, Yijia Zhang","doi":"10.1109/TCBBIO.2026.3746316","DOIUrl":"https://doi.org/10.1109/TCBBIO.2026.3746316","url":null,"abstract":"Computational drug repositioning (DR) discovers novel therapeutic applications for existing pharmaceuticals, offering an efficient alternative to costly de novo development. Recently, to capture higher-order biological interactions, hypergraph-based DR approaches are devised. Nevertheless, this line of research encounters two fundamental bottlenecks that compromise their real-world practicality. (i) Insufficient biological mediator integration. The hypergraph construction schemes neglect diverse biological intermediates that mediate drug-disease therapeutic mechanisms, instead relying solely on drug and disease similarity networks or drug-disease association (DDA) bipartite graph. (ii) Unresolved data sparsity challenge. The limited confirmed DDAs compel hypergraph-based approaches to perform supervised learning on sparse positive instances. In this study, we propose a novel hypergraph masked autoencoder model for DR, namely Hyades. To handle the first bottleneck, we retrieve biological entities that maintain direct interaction relationships with drugs and diseases from three large-scale biomedical knowledge graphs. Hence, genes, as the exclusive biological mediators, are conceptualized as hyperedges connecting corresponding drug and disease nodes. For the second bottleneck, we resort to integrating the graph masked autoencoder with hypergraph neural network for generating complementary self-supervised signals through masked node feature reconstruction. Extensive experiments across three DR datasets under 10-fold cross-validation confirm the model effectiveness.","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149248478","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}

引用次数: 0

HetAnn: Cell Type Annotation for Spatial Transcriptomics by Heterogeneous Graph Learning. 基于异构图学习的空间转录组学细胞类型标注。

IEEE transactions on computational biology and bioinformatics Pub Date : 2026-10-21 DOI: 10.1109/TCBBIO.2026.3745540

Yang Zhang, Guannan Du, Zhiming Dai

{"title":"HetAnn: Cell Type Annotation for Spatial Transcriptomics by Heterogeneous Graph Learning.","authors":"Yang Zhang, Guannan Du, Zhiming Dai","doi":"10.1109/TCBBIO.2026.3745540","DOIUrl":"https://doi.org/10.1109/TCBBIO.2026.3745540","url":null,"abstract":"Currently, several new spatial transcriptomics(ST) technologies have attained cellular or even subcellular spatial resolution. Effective methods for annotating cell types are essential for the growing volumes of single-cell ST data. Most of automatic annotation methods for ST data use reference single cell RNA-sequencing(scRNA-seq) data, but how to efficiently integrate spatial information from ST data and gene expression information from ST and scRNA-seq data into a unified model remains a challenge. The difficulty in integrating the single cell multimodal data is that different modalities have different feature spaces, which easily leads to cell type mismatching. In this study, we propose HetAnn, a semi-supervised deep learning method for cell type annotation of ST data. HetAnn learns ST-specific representation by cell graphs accounting for both cell-to-cell spatial distance and gene expression similarity in ST data. HetAnn employs a graph attention network to learn multimodal representation with end-to-end joint training on ST and reference scRNA-seq data. These two comprehensive embeddings enable HetAnn to remove the gap between modalities while preserving cell-type heterogeneity. Experiments on public datasets demonstrate the superiority of HetAnn compared with state-of-the-art methods.","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149241165","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}

引用次数: 0

A Biological Property-Aware Quantum-Classical Fusion Model for Predicting AAV Capsid Production Fitness. 预测AAV衣壳生产适应度的生物特性感知量子经典融合模型。

IEEE transactions on computational biology and bioinformatics Pub Date : 2026-10-20 DOI: 10.1109/TCBBIO.2026.3745380

Lu Wang, Fanxu Meng, Yuxiang Liu, Zaichen Zhang, Xutao Yu

{"title":"A Biological Property-Aware Quantum-Classical Fusion Model for Predicting AAV Capsid Production Fitness.","authors":"Lu Wang, Fanxu Meng, Yuxiang Liu, Zaichen Zhang, Xutao Yu","doi":"10.1109/TCBBIO.2026.3745380","DOIUrl":"10.1109/TCBBIO.2026.3745380","url":null,"abstract":"Accurate prediction of capsid production fitness remains a major challenge in the rational design of adeno-associated virus (AAV) variants, owing to the highly nonlinear and heterogeneous relationship between amino acid sequence patterns and production outcomes. To address this problem, we propose a biological property-aware quantum-classical adaptive fusion model for predicting AAV capsid production fitness directly from amino acid sequences. The model represents sequences using biologically meaningful physicochemical properties and introduces a dual-signal adaptive controller that generates two task-dependent latent signals for regulating quantum encoding and entanglement operations. Based on these learned signals, the model dynamically adjusts quantum feature mapping and quantum-classical feature fusion, enabling sample-dependent regulation of quantum processing and heterogeneous feature integration. Experimental results over five independent runs show that the proposed model achieves a Pearson correlation coefficient of $0.919pm 0.003$ and an MAE of $0.38pm 0.02$, outperforming the traditional machine learning and general neural network baselines while using only 0.14M trainable parameters. Compared with the strongest task-relevant baseline, it achieves comparable Pearson correlation performance and a statistically supported reduction in MAE, demonstrating a favorable predictive-performance-parameter-efficiency trade-off. The results suggest that biologically informed quantum-classical hybrid learning is a promising direction for predictive modeling in AAV capsid engineering.","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149229862","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}

引用次数: 0

Inferring gene regulatory networks from single-cell transcriptomics jointly through graph neural networks and Transformer. 通过图神经网络和Transformer共同推断单细胞转录组学的基因调控网络。

IEEE transactions on computational biology and bioinformatics Pub Date : 2026-10-16 DOI: 10.1109/TCBBIO.2026.3740828

Yanglan Gan, Tao Liu, Guangwei Xu, Wenjing Guo, Guobing Zou

{"title":"Inferring gene regulatory networks from single-cell transcriptomics jointly through graph neural networks and Transformer.","authors":"Yanglan Gan, Tao Liu, Guangwei Xu, Wenjing Guo, Guobing Zou","doi":"10.1109/TCBBIO.2026.3740828","DOIUrl":"10.1109/TCBBIO.2026.3740828","url":null,"abstract":"Inferring gene regulatory networks (GRNs) is crucial for understanding the complex regulatory relationships that govern gene expression in cells. Recently, several computational methods have been proposed for reconstructing GRNs from single-cell RNA sequencing (scRNA-seq) data. However, most of these methods primarily focus on pairwise regulatory relationships between genes, failing to leverage the global regulatory structure of GRNs, which is crucial for exploring the complex regulatory interactions between transcription factors (TFs) and their target genes. To address these limitations, we propose GTGRNS, a supervised learning framework that integrates a graph neural network with a Transformer encoder for GRN inference from scRNA-seq data. Specifically, we utilize the GraphSAGE model to integrate regulatory network topological information with gene expression features, facilitating the learning of gene representations. By leveraging the Transformer model, we further extract complex dependencies of gene pairs, which are crucial for determining regulatory levels. The comparative experiments on several benchmark datasets demonstrate the effectiveness and competitiveness of the proposed GTGRNS in the task of GRN inference.","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149208247","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}

引用次数: 0

Hypergraph-Enhanced Self-Supervised Learning with Module-Aware Structured Masking for Single-Cell Transcriptomic Clustering. 单细胞转录组聚类的超图增强自监督学习与模块感知结构化掩蔽。

IEEE transactions on computational biology and bioinformatics Pub Date : 2026-10-14 DOI: 10.1109/TCBBIO.2026.3743566

Yaowei Cao, Xinyu Du, Zhenhua Yu

{"title":"Hypergraph-Enhanced Self-Supervised Learning with Module-Aware Structured Masking for Single-Cell Transcriptomic Clustering.","authors":"Yaowei Cao, Xinyu Du, Zhenhua Yu","doi":"10.1109/TCBBIO.2026.3743566","DOIUrl":"10.1109/TCBBIO.2026.3743566","url":null,"abstract":"Accurate clustering of single-cell RNA sequencing (scRNA-seq) data is essential for cell type annotation and cellular heterogeneity inference, but existing methods are still limited by local gene co-expression redundancy and pairwise graph modeling, which hinder the capture of cross-module biological semantics and higher-order topological cellular relationships. To address these challenges, we propose a novel method called scHyperMod. It employs a module-aware structured masking strategy that suppresses shortcut learning driven by local co-expression redundancy and encourages the model to learn more discriminative representations from cross-module contexts. Furthermore, hypergraph modeling based on shared nearest neighbors is introduced to enable information propagation across locally connected cells, thus providing a more expressive description of higher-order topology. Extensive evaluations on 12 public benchmark datasets show that scHyperMod achieves leading overall clustering performance across varying data scales, sparsity levels, and cellular heterogeneity. Application to a glioblastoma dataset further demonstrates its ability to recover major cell populations and distinguish malignant cell states, including AC-like, OPC-like, and Cycling, while preserving cell identity-related biological differences in marker gene expression and functional enrichment.","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149187366","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}

引用次数: 0

Multi-Objective Molecular Generation via Actor-Critic Regulated by Adaptive Cooperative Evolution. 自适应协同进化调控的多目标分子生成。

IEEE transactions on computational biology and bioinformatics Pub Date : 2026-10-14 DOI: 10.1109/TCBBIO.2026.3743558

Haoran Yang, Fei Zhu

{"title":"Multi-Objective Molecular Generation via Actor-Critic Regulated by Adaptive Cooperative Evolution.","authors":"Haoran Yang, Fei Zhu","doi":"10.1109/TCBBIO.2026.3743558","DOIUrl":"10.1109/TCBBIO.2026.3743558","url":null,"abstract":"Many complex diseases involve intricate and interdependent biological mechanisms, making multi-target drug design therapeutically important. However, existing molecular generation models often struggle to balance conflicting property requirements, maintain structural exploration, and avoid premature convergence during reinforcement learning. To address these challenges, we propose CMIR-RL, a reinforcement learning-based molecular generation model that integrates cooperative crossover-mutation exploration, structure-based intrinsic rewards, and Pareto-based multi-objective optimization. In CMIR-RL, mutation and crossover networks synergistically guide the agent to balance exploration and exploitation, while Pareto-front-guided reward assignment updates the generation policy under multiple property constraints. Experiments on a three-objective polypharmacology task show that CMIR-RL significantly improves desirability from 0.8868 to 0.9542 over the baseline method, while maintaining competitive validity, uniqueness, and diversity. Additional GuacaMol results further indicate competitive generalization across goal-directed molecular generation tasks. These results show that CMIR-RL is an effective reinforcement learning strategy for early-stage multi-objective molecular design.","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149187527","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}

引用次数: 0

An Efficient and Exact Algorithm for Scaffolding Contigs Using Exemplar Double-Cut-and-Join Distance. 基于样例双切割连接距离的高效、精确的脚手架组合算法。

IEEE transactions on computational biology and bioinformatics Pub Date : 2026-10-13 DOI: 10.1109/TCBBIO.2026.3742524

Ming-Yeh Chou, Wen-Rui Liao, Chin Lung Lu

{"title":"An Efficient and Exact Algorithm for Scaffolding Contigs Using Exemplar Double-Cut-and-Join Distance.","authors":"Ming-Yeh Chou, Wen-Rui Liao, Chin Lung Lu","doi":"10.1109/TCBBIO.2026.3742524","DOIUrl":"https://doi.org/10.1109/TCBBIO.2026.3742524","url":null,"abstract":"Reference-based scaffolding is an important process in chromosome-scale genome sequencing, which is to utilize a reference genome to scaffold (i.e., order and orient) the contigs generated by assembly tools for a target genome. The reference-based scaffolding was formulated as a rearrangement distance-based scaffolding problem, which is to scaffold the contigs in two given draft genomes, both containing duplicate markers and acting as references to each other, such that the rearrangement distance between the scaffolded genomes is minimized. In this work, we study this problem by considering exemplar double-cut-and-join (DCJ) distance. We show that the exemplar DCJ distance-based scaffolding problem is NP-hard and design an efficient integer linear programming (ILP)-based algorithm (denoted by EDCJ-scaffolder) to exactly solve it. We also use simulated and biological datasets to test EDCJ-scaffolder and compare its accuracy with that of another ILP-based scaffolding algorithm (denoted by EBP-scaffolder) we previously designed to solve the exemplar breakpoint distance-based scaffolding problem. Our experimental results finally show that EDCJ-scaffolder can more accurately scaffold the contigs of a draft genome as compared to EBP-scaffolder. Particularly, the sensitivity of EBP-scaffolder decreases as the rearrangement distance between target and reference genomes increases in simulated datasets, while the sensitivity of EDCJ-scaffolder remains high consistently.","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149187325","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}

引用次数: 0