IEEE transactions on computational biology and bioinformatics最新文献_第4页

DMGRN: Enhancing Diffusion Models for Gene Regulatory Network Inference. DMGRN：基因调控网络推断的增强扩散模型。

IEEE transactions on computational biology and bioinformatics Pub Date : 2026-09-10 DOI: 10.1109/TCBBIO.2026.3731227

Rongyuan Li, Jingli Wu, Chunfeng Chen, Gaoshi Li, Jiafei Liu, Haize Hu, Junbo Xuan, Jinlu Liu, Zheng Deng, Daoqing Gong

{"title":"DMGRN: Enhancing Diffusion Models for Gene Regulatory Network Inference.","authors":"Rongyuan Li, Jingli Wu, Chunfeng Chen, Gaoshi Li, Jiafei Liu, Haize Hu, Junbo Xuan, Jinlu Liu, Zheng Deng, Daoqing Gong","doi":"10.1109/TCBBIO.2026.3731227","DOIUrl":"10.1109/TCBBIO.2026.3731227","url":null,"abstract":"Gene regulatory networks (GRNs) encode the intricate interactions between transcription factors (TFs) and their target genes, playing a pivotal role in orchestrating cellular metabolism, proliferation, and differentiation, thereby illuminating the molecular mechanisms underlying disease onset and progression. The increasing availability of single-cell RNA sequencing (scRNA-seq) data offers unprecedented opportunities for computational GRN inference. However, the inherent high noise and sparsity of scRNA-seq data considerably impair the performance of existing inference methods. To overcome these limitations, we propose DMGRN, a novel GRN inference frame work built upon an improved Denoising Diffusion Probabilistic Model (DDPM). Our approach first applies a forward diffusion process to progressively introduce Gaussian noise into the raw expression data, and subsequently employs a reverse process integrated with a structural equation model (SEM) to predict the noise, thereby accurately recovering gene regulatory relationships. To further enhance inference fidelity, we introduce a gene similarity alignment loss that encourages correlation consistency between the predicted noise and the perturbed data at the gene level, enabling the model to simultaneously capture cellular-level noise residuals and gene-level regulatory co-expression patterns. Experimental results on 28 BEELINE benchmark datasets demonstrate that DMGRN achieves the highest Early Precision Ratio (EPR) on 18 out of 28 BEELINE benchmark configurations, demonstrating superior stability and computational efficiency compared to state-of-the-art methods.","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148932328","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

ScGraphTrans: Pathway-Guided Graph Learning and Domain Adaptation for Cell Type Annotation in Single-Cell RNA-seq. 单细胞RNA-seq中细胞类型标注的路径引导图学习和域适应。

IEEE transactions on computational biology and bioinformatics Pub Date : 2026-09-10 DOI: 10.1109/TCBBIO.2026.3733126

Yue-Chao Li, Hai-Ru You, Meng-Meng Wei, Xin-Fei Wang, Yu Li, Zhi-An Huang, Yu-An Huang, Zhu-Hong You

{"title":"ScGraphTrans: Pathway-Guided Graph Learning and Domain Adaptation for Cell Type Annotation in Single-Cell RNA-seq.","authors":"Yue-Chao Li, Hai-Ru You, Meng-Meng Wei, Xin-Fei Wang, Yu Li, Zhi-An Huang, Yu-An Huang, Zhu-Hong You","doi":"10.1109/TCBBIO.2026.3733126","DOIUrl":"10.1109/TCBBIO.2026.3733126","url":null,"abstract":"The tumor microenvironment (TME) is a complex ecosystem in which intercellular communication regulates tumor progression and therapeutic response. Yet inferring cell-cell interactions from non-spatial scRNA-seq remains challenging due to incomplete ligand-receptor databases and inaccurate cell type annotations. Here, we propose scGraphTrans, a graph neural network framework that integrates functional state pseudo-labels, graph structure learning, and graph domain adaptation to improve both cell type annotation and communication inference. Pathway activity scores across 14 cancer-relevant processes (e.g., angiogenesis, apoptosis, cell cycle) are used as pseudo-labels to refine cell-cell graphs, capturing functional proximity beyond geometric similarity. A domain adaptation module further aligns embeddings across patients, enhancing cross-individual generalization. Evaluated on 38,667 cells from 15 individuals across three cancers, scGraphTrans achieved an average accuracy of 84.28%, surpassing state-of-the-art baselines while maintaining robustness across heterogeneous datasets. Statistical validation demonstrated recovery of disease-specific gene interactions (e.g., LGALS1-SUSD2 in breast invasive carcinoma and BIRC5-CASP6 in colorectal cancer) without prior ligand-receptor supervision. The source code and data used in this paper can be found in https://github.com/LiYuechao1998/scGraphTrans. Our framework thus provides an interpretable and generalizable solution for TME analysis, offering insights into biomarker discovery and therapeutic strategies.","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148932316","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

MV-MLCF: A Semi-Supervised Self-training Method for Peptide Function Prediction Based on Multi-View and Multi-Label Co-Forest. MV-MLCF：一种基于多视图和多标签协同森林的半监督自训练多肽函数预测方法。

IEEE transactions on computational biology and bioinformatics Pub Date : 2026-09-08 DOI: 10.1109/TCBBIO.2026.3732203

Jicong Duan, Shang Zheng, Xibei Yang, Hualong Yu

{"title":"MV-MLCF: A Semi-Supervised Self-training Method for Peptide Function Prediction Based on Multi-View and Multi-Label Co-Forest.","authors":"Jicong Duan, Shang Zheng, Xibei Yang, Hualong Yu","doi":"10.1109/TCBBIO.2026.3732203","DOIUrl":"10.1109/TCBBIO.2026.3732203","url":null,"abstract":"Peptides, as biologically active molecules, have gained significant attention due to their therapeutic potential in treating various diseases, including metabolic disorders, cancers, and infectious diseases. However, the identification and functional annotation of multifunctional therapeutic peptides (MTPs) still remain challenging due to the limitations of traditional experimental costs and the scarcity of high-quality labeled data. To address these issues, we propose a novel semi-supervised self-training method called MV-MLCF (Multi-View and Multi-Label Co-Forest). MV-MLCF leverages multi-view feature representation techniques to comprehensively represent peptide sequences and employs a co-forest framework to iteratively label unannotated instances with a small set of high-quality labeled samples. By integrating multi-label learning, MV-MLCF effectively captures the multifunctional nature of peptides and lowers dependency on extensive labeled datasets. Extensive experiments across three benchmark datasets demonstrate the robustness and generalization of MV-MLCF across multiple base classifiers, including logistic regression, naive Bayes, and extreme learning machines. The results show that the proposed MV MLCF method significantly improves prediction accuracy and outperforms comparison algorithms, highlighting its potential in accelerating peptide-based drug discovery and reducing experimental costs. This study provides a new perspective and methodology for peptide function prediction, and offers a scalable and efficient solution to overcome data scarcity and enhance prediction performance in computational biology and bioinformatics.","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148915036","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

Graph-Attentional Deep Sparse Subspace Clustering for Single-Cell Transcriptomics. 单细胞转录组学的图注意深度稀疏子空间聚类。

IEEE transactions on computational biology and bioinformatics Pub Date : 2026-09-07 DOI: 10.1109/TCBBIO.2026.3731206

Jingli Wu, Shihao Zhang, Gaoshi Li, Xiaopeng Wei, Jiafei Liu, Haize Hu

{"title":"Graph-Attentional Deep Sparse Subspace Clustering for Single-Cell Transcriptomics.","authors":"Jingli Wu, Shihao Zhang, Gaoshi Li, Xiaopeng Wei, Jiafei Liu, Haize Hu","doi":"10.1109/TCBBIO.2026.3731206","DOIUrl":"10.1109/TCBBIO.2026.3731206","url":null,"abstract":"Accurate identification of cell types constitutes a critical step in downstream analysis of single-cell sequencing data. However, the inherent high noise levels and high dimensionality characteristics pose significant challenges for clustering tasks. To address these issues, we propose method scGADSSC (Graph-Attentional Deep Sparse Subspace Clustering for Single-Cell Transcriptomics), an innovative end-to-end framework that achieves joint optimization and mutual enhancement of graph attention learning and subspace self-representation. It consists of two core collaborative components: (1) A Denoising Autoencoder for explicit modeling and noise reduction of raw expression data; (2) A Graph Attention Autoencoder to learn a discriminative self-expression matrix, which is subsequently used to construct a similarity matrix for spectral clustering-based cell type classification. Comprehensive evaluations across 15 biological datasets demonstrate that scGADSSC outperforms ten state-of-the-art single-cell clustering methods on most test datasets, achieving superior clustering performance.","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148908628","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

KEPLA: A Knowledge-Enhanced Deep Learning Framework for Accurate Protein-Ligand Binding Affinity Prediction. KEPLA：一个知识增强的深度学习框架，用于准确预测蛋白质-配体结合亲和力。

IEEE transactions on computational biology and bioinformatics Pub Date : 2026-09-01 DOI: 10.1109/TCBBIO.2026.3715687

Han Liu, Keyan Ding, Peilin Chen, Yinwei Wei, Liqiang Nie, Dapeng Wu, Shiqi Wang

{"title":"KEPLA: A Knowledge-Enhanced Deep Learning Framework for Accurate Protein-Ligand Binding Affinity Prediction.","authors":"Han Liu, Keyan Ding, Peilin Chen, Yinwei Wei, Liqiang Nie, Dapeng Wu, Shiqi Wang","doi":"10.1109/TCBBIO.2026.3715687","DOIUrl":"10.1109/TCBBIO.2026.3715687","url":null,"abstract":"Accurate prediction of protein-ligand binding affinity is critical for drug discovery. While recent deep learning approaches have demonstrated promising results, they often rely solely on structural features of proteins and ligands, overlooking their valuable biochemical knowledge associated with binding affinity. To address this limitation, we propose KEPLA, a novel deep learning framework that explicitly integrates prior knowledge from Gene Ontology and ligand properties to enhance prediction performance. KEPLA takes protein sequences and ligand molecular graphs as input and optimizes two complementary objectives: 1) aligning global representations with knowledge graph relations to capture domain-specific biochemical insights, and 2) leveraging cross attention between local representations to construct fine-grained joint embeddings for prediction. Experiments on two benchmark datasets across both in-domain and cross-domain scenarios demonstrate that KEPLA consistently outperforms state-of-the-art baselines. Furthermore, interpretability analyses based on knowledge graph relations and cross attention maps provide valuable insights into the underlying predictive mechanisms.","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":"2038-2051"},"PeriodicalIF":0.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148552067","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

CA-GCN: A Causality-Aware Graph Coupling Network for Multimodal Learning of Brain Connectivity. CA-GCN：一种用于大脑连接多模态学习的因果感知图耦合网络。

IEEE transactions on computational biology and bioinformatics Pub Date : 2026-09-01 DOI: 10.1109/TCBBIO.2026.3725309

Zhiwei Song, Zhengyuan Lyu, Si Fu, Hu Yu, Xiaojuan Guo

{"title":"CA-GCN: A Causality-Aware Graph Coupling Network for Multimodal Learning of Brain Connectivity.","authors":"Zhiwei Song, Zhengyuan Lyu, Si Fu, Hu Yu, Xiaojuan Guo","doi":"10.1109/TCBBIO.2026.3725309","DOIUrl":"10.1109/TCBBIO.2026.3725309","url":null,"abstract":"Understanding the coupling relationship between structural connectivity (SC) and functional connectivity (FC) is essential for advancing our knowledge of individual brain function and organization. Recent deep learning techniques, particularly graph neural networks (GNNs), have shown strong potential in modeling SC-FC coupling relationships. However, existing multimodal graph-learning methods often overlook intrinsic causal relationships and task-specific SC-FC coupling patterns, thereby limiting both predictive performance and interpretability. To address these issues, we propose a Causality-Aware Graph Coupling Network (CA-GCN) that disentangles causal representations and models causal SC-FC coupling patterns in a task-relevant manner. Specifically, we first develop a Multimodal Causal Disentanglement (MCD) module that isolates SC and FC causal features related to the task label via Granger causality-inspired learning. Then, we design a Multimodal Causal Coupling (MCC) module that identifies SC-FC coupling patterns through contrastive learning of cross-modal attention. Evaluated on HCPYA for fluid cognition prediction and CamCAN for brain age prediction, CA-GCN consistently outperforms single-modal and state-of-the-art multimodal graph methods. Moreover, the interpretable saliency maps of CA-GCN reveal the key causal brain regions and SC-FC coupling patterns.","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":"2244-2256"},"PeriodicalIF":0.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148805427","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

BIKE: A Binary $K$-Mer Exact Counter With Alphabet-Independent Memory and Deterministic Parallelism. 具有字母无关存储器和确定性并行性的二元K -mer精确计数器。

IEEE transactions on computational biology and bioinformatics Pub Date : 2026-09-01 DOI: 10.1109/TCBBIO.2026.3723002

Fabio F de Oliveira, Marcelo A C Fernandes

{"title":"BIKE: A Binary $K$-Mer Exact Counter With Alphabet-Independent Memory and Deterministic Parallelism.","authors":"Fabio F de Oliveira, Marcelo A C Fernandes","doi":"10.1109/TCBBIO.2026.3723002","DOIUrl":"10.1109/TCBBIO.2026.3723002","url":null,"abstract":"$K$-mer counting is a fundamental computational task in bioinformatics, underpinning genome assembly, metagenomic classification, error correction, and similarity analysis. Existing exact-counting methods rely on hash tables or static allocation strategies whose memory requirements grow exponentially with the alphabet size and substring length, rendering them impractical for amino acid sequences at moderate-to-large values of $k$. We propose BIKE (Binary $K$-mer Exact Counter), a novel exact $k$-mer counting algorithm whose memory footprint depends exclusively on the input sequence length $n$, independently of the alphabet size $m$ or the $k$-mer length $k$. BIKE decomposes the counting problem into $n-1$ mutually independent pivot-based comparison blocks operating entirely on binary matrices, requiring only one bit per entry, and employs a union-find aggregation mechanism that guarantees exact counts for $k$-mers of arbitrary multiplicity. This structural regularity yields a fully deterministic degree of parallelism, enabling closed-form analytical models that provide accurate execution-time predictions under ideal parallel execution assumptions. Experimental results on real biological sequences confirm functional correctness and demonstrate memory reductions of up to three orders of magnitude over classical exact methods for amino acid alphabets. Analytical performance projections, derived from the closed-form parallel model, indicate that an FPGA realisation of BIKE would be expected to outperform CPU-based dynamic allocation at moderate sequence lengths; however, these remain theoretical estimates pending hardware implementation. BIKE is therefore presented as a theoretical and data-structural contribution, establishing a new algorithmic foundation for alphabet-independent, exactly-counted, and deterministically parallel $k$-mer analysis.","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":"2181-2192"},"PeriodicalIF":0.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148727861","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

Quantum-Enhanced Transfer Learning for IDR Binding Partner Prediction. 基于量子增强迁移学习的IDR结合伙伴预测。

IEEE transactions on computational biology and bioinformatics Pub Date : 2026-09-01 DOI: 10.1109/TCBBIO.2026.3724278

Seok-Jin Kang, Hongchul Shin

{"title":"Quantum-Enhanced Transfer Learning for IDR Binding Partner Prediction.","authors":"Seok-Jin Kang, Hongchul Shin","doi":"10.1109/TCBBIO.2026.3724278","DOIUrl":"10.1109/TCBBIO.2026.3724278","url":null,"abstract":"Intrinsically Disordered Regions (IDRs) play essential roles in cellular processes through interactions with proteins, nucleic acids, lipids, and metal ions, yet predicting their binding partners remains challenging for understanding protein function and drug discovery. However, current computational methods including protein language models face performance plateaus where traditional approaches to improve accuracy have become ineffective. Here, we present a hybrid quantum-classical machine learning approach that combines variational quantum circuits with the ESM2 protein language model for multi-class IDR binding partner prediction using a prototypical network. Through systematic evaluation of quantum circuit architectures across factorial experiments, we demonstrate that the hybrid model achieves statistically significant performance improvements over classical baselines, with entanglement topology governing model stability and encoding methods determining performance gains. These findings establish that quantum advantage in computational biology emerges from architectural design principles rather than computational scale, providing a framework for overcoming performance limitations in bioinformatics applications where dataset expansion is constrained.","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":"2234-2243"},"PeriodicalIF":0.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148805458","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

Boundary-Aware Clustering of Spatial Transcriptomics Data via Fourier Feature Mapping and Dynamic Self-Supervision. 基于傅立叶特征映射和动态自我监督的空间转录组学数据的边界感知聚类。

IEEE transactions on computational biology and bioinformatics Pub Date : 2026-09-01 DOI: 10.1109/TCBBIO.2026.3710630

Bohua Li, Yanran Zhu, Xiao Zheng, Hengjian Tong, Yuanyuan Liu, Chang Tang, Xinwang Liu, Kunlun He

{"title":"Boundary-Aware Clustering of Spatial Transcriptomics Data via Fourier Feature Mapping and Dynamic Self-Supervision.","authors":"Bohua Li, Yanran Zhu, Xiao Zheng, Hengjian Tong, Yuanyuan Liu, Chang Tang, Xinwang Liu, Kunlun He","doi":"10.1109/TCBBIO.2026.3710630","DOIUrl":"10.1109/TCBBIO.2026.3710630","url":null,"abstract":"Accurate identification of fine-grained spatial domains within tissues using spatial transcriptomics data is critical for understanding tissue development, disease pathology, and cellular interactions. However, inherent spatial heterogeneity in spatial transcriptomic data, characterized by significant discontinuity in spatial gradients of gene expression, makes it difficult for existing clustering methods to precisely delineate fine-grained structures and produce ambiguous spatial boundaries. To address this challenge, this paper proposes FPS-MGCN - a multi-view graph convolutional network integrating Fourier feature mapping with dynamic pseudo-label self-supervision. In detail, to tackle boundary ambiguity caused by spatial heterogeneity, we introduce a Fourier sine-cosine feature mapping that explicitly constructs high-frequency feature channels to enhance sensitivity to high-frequency spatial gradients such as tissue boundaries. In addition, a dynamic pseudo-label self-supervision strategy is further designed, providing progressive semantic constraints that smooth noise interference and enhance recognition in spatial boundary regions. The model integrates raw gene expression, Fourier-mapped features, and spatial adjacency information through a multi-view GCN encoder, employing an attention mechanism for adaptive fusion. Extensive experimental results demonstrate that the proposed FPS-MGCN identifies critical fine-grained structures difficult to resolve by traditional methods and outperforms other state-of-the-art methods in spatial domain identification.","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":"1928-1940"},"PeriodicalIF":0.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148407918","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

Predicting miRNA-Disease Associations Based on Adaptive Neighborhood Propagation and Feature Spatial Recombination. 基于适应性邻域传播和特征空间重组的mirna -疾病关联预测。

IEEE transactions on computational biology and bioinformatics Pub Date : 2026-09-01 DOI: 10.1109/TCBBIO.2026.3715668

Ru Nie, Yingkai Li, Zhengwei Li, Lei Wang, Zhanguo Xia, Yujun Zhao

{"title":"Predicting miRNA-Disease Associations Based on Adaptive Neighborhood Propagation and Feature Spatial Recombination.","authors":"Ru Nie, Yingkai Li, Zhengwei Li, Lei Wang, Zhanguo Xia, Yujun Zhao","doi":"10.1109/TCBBIO.2026.3715668","DOIUrl":"10.1109/TCBBIO.2026.3715668","url":null,"abstract":"MicroRNAs (miRNAs) are critical regulators in biological processes such as cell proliferation, differentiation, and apoptosis, with their aberrant expression strongly linked to a range of complex diseases. Because traditional experimental methods for predicting miRNA-disease associations (MDAs) are both time-intensive and costly, computational models offer an efficient alternative. Graph neural networks (GNNs) have shown promise in MDAs prediction. However, existing models often suffer from limitations, including inadequate neighborhood information aggregation, inflexible propagation schemes, and an imbalance between global and local information. To address these issues, this paper presents a novel GNN framework, APKAGN, designed for predicting miRNA-disease associations. APKAGN enhances performance through three innovative mechanisms: 1) Adaptive local propagation, leveraging a gated recursion module to dynamically adjust propagation depth while employing residual connections to preserve multi-scale features. 2) Multi-subspace global aggregation, capturing global topology information via multi-dimensional projection and density-aware KNN selection. 3) Dynamic feature fusion, integrating local and global representations using an attention-based gating mechanism. Evaluated on the updated HMDD v3.2 dataset across multiple independent random seeds, APKAGN achieved an outstanding average AUC of 95.09%, an accuracy of 88.23%, and an F1-score of 88.34%, outperforming seven state-of-the-art baseline models. Case studies on lymphoma, prostate, and breast tumors further demonstrated the predictive performance of the proposed model, with 26, 25, and 27 of the top 30 predicted miRNAs validated in the dbDEMC and miR2Disease databases, respectively. By leveraging adaptive propagation and dynamic KNN mechanisms, APKAGN significantly enhances the accuracy of MDAs prediction, offering a powerful tool for investigating disease mechanisms and identifying biomarkers.","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":"2066-2077"},"PeriodicalIF":0.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148552108","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