IEEE transactions on computational biology and bioinformatics最新文献

{"title":"Minimum Description Length-Driven Fragment Mining for Pretraining Molecule Property Prediction Model.","authors":"Xing Cai, Tong Zhang, Yide Qiu, Xudong Wang, Baotong Su, Zhen Cui","doi":"10.1109/TCBBIO.2026.3697250","DOIUrl":"10.1109/TCBBIO.2026.3697250","url":null,"abstract":"<p><p>Molecular fragments play a crucial role in molecular property prediction. However, most existing deep learning approaches rely heavily on expert-defined substructural patterns, limiting their ability to identify novel or latent fragments. This constraint reduces the generalizability and applicability of molecular fragments in molecular representation learning. In this study, we propose the Molecular Multi-view Pre-training model with Adaptive Fragment Mining (MMP-AFM), a unified framework that facilitates the seamless integration of molecular structural information. MMP-AFM formulates fragment discovery as a combinatorial optimization problem, using description length as the objective to enable adaptive extraction of molecular fragments and dynamic construction of a fragment library. Additionally, we design a molecular multi-view self-supervised pretraining framework that aligns features from the fragment, global, and data augmentation views, ensuring a comprehensive integration of molecular substructural information. Finally, the MMP-AFM is applied to both molecular classification and regression tasks. Experimental results demonstrate that MMP-AFM consistently outperforms existing methods across multiple tasks, highlighting its broad applicability and efficiency.</p>","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":"1747-1760"},"PeriodicalIF":0.0,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148039270","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 Realistic Control Approach for Set Stabilization of Complex Biological Networks With Logical Models.","authors":"Jung-Min Yang, Namhee Kim, Kwang-Hyun Cho","doi":"10.1109/TCBBIO.2026.3692800","DOIUrl":"10.1109/TCBBIO.2026.3692800","url":null,"abstract":"<p><p>Driving cellular biological systems to one of desired long-term behaviors, an important subject in systems biology, can be interpreted as the set stabilization problem. This article presents a realistic control approach to set stabilization of complex biological networks modeled by Boolean networks (BNs) wherein only open-loop control is applied. First, the considered BN is transformed into an auxiliary BN having redundant state variables. Based on the similarity between desired attractors and Boolean algebra, a number of reduced BNs are constructed from the auxiliary BN in favor of set stabilization. Next, the feedback vertex set (FVS) control law is applied to reduced BNs for obtaining open-loop control inputs that take partial state variables toward desired values. This procedure continues in a recursive way until ensuring that the controlled BN converges to an unspecified desired attractor. The proposed scheme has the superiority that both fixed-point and cyclic attractors can be included in the problem setting, and that it can tackle complex BNs with moderate computational burdens. Extensive numerical experiments on random BNs and real biological systems are provided to demonstrate the applicability of the proposed scheme.</p>","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":"1656-1670"},"PeriodicalIF":0.0,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147937240","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 Multi-Branch Feature Fusion Transformer Network and Its Application in Neck Ultrasound Detection.","authors":"Qing Guo, Jie Yang, Ming-An Yu, Xiaoyu Yang, Ying Wei, Zhenlong Zhao, Huaqing Wang","doi":"10.1109/TCBBIO.2026.3692793","DOIUrl":"10.1109/TCBBIO.2026.3692793","url":null,"abstract":"<p><p>Hyperparathyroidism (HPT) and thyroid nodule (TN) are caused by abnormalities in the parath/yroid and thyroid glands, respectively. Due to their proximity, small size, and similar ultrasound characteristics, the traditional object detection algorithms often struggle to accurately differentiate between HPT and TN when both lesions coexist in ultrasound images, leading to a high rate of misdiagnosis. In order to achieve accurate detection of HPT and TN when the two lesions coexist, we constructed three comprehensive object detection datasets: one containing only hyperparathyroidism (HPTD), one containing only thyroid nodules (TND), and one mixed dataset that includes both types of lesions (HPT-TND). A novel multi-branch feature fusion DETR network (MB-DETR) is proposed based on the Real-Time Detection Transformer (RT-DETR) model. We redesigned the feature fusion module and incorporated asymmetric convolution to enhance feature extraction. To validate the proposed MB-DETR performance, the experiments have been carried out on the three datasets. Our model achieved a superior performance compared to the state-of-the-art object detection models in the key metrics such as F1, Precision, and Recall, while significantly reducing computational costs. Additionally, the ablation studies confirmed the effectiveness of asymmetric convolution and the multi-branch feature fusion module in terms of enhancement of detection performance. The experimental results show that the Multi-Branch Feature Fusion incorporated with the asymmetric convolution improves the local feature extraction capability of the DETR model. It is concluded that the proposed MB-DETR model outperforms the existing ones in the detection of TN and HPT when both lesions coexist and thus effectively assists in the diagnosis of the correlated disease.</p>","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":"1574-1584"},"PeriodicalIF":0.0,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147937242","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":"EssentCell: Discovering Essential Evolutionary Relations in Noisy Single-Cell Data.","authors":"Adiesha Liyanage, Robyn Burger, Allison Shi, Braeden Sopp, Binhai Zhu, Brendan Mumey","doi":"10.1109/TCBBIO.2026.3691222","DOIUrl":"10.1109/TCBBIO.2026.3691222","url":null,"abstract":"<p><p>Single-cell sequencing (SCS) enables the study of tumor evolution at the resolution of a single cell. SCS data can be represented as a binary matrix, where the $ij$-th entry indicates whether cell $i$ has mutation $j$. There is a simple characterization of when the data is compatible with a perfect phylogeny based on the absence of a special \"conflict\" submatrix. In practice, SCS data are noisy, which raises the natural question of the minimum number of entries that must be flipped in the data matrix to make it conflict-free and thus compatible with a perfect phylogeny. Furthermore, the likelihood of a false positive is several orders of magnitude smaller than that of a false negative rate. We consider a variation of the minimum-flip problem parameterized by the number of false positives. Restricting the false positive rate to a small range, often multiple optimal solutions can arise. While previous work has focused on reconstructing a single optimal phylogenetic tree, we are interested in the relations that are present among all optimal solutions; we call such relations essential. In this work, we propose an efficient algorithm based on integer linear programming to determine the essential relation on the cells given an SCS data matrix. We test our tool, ${sf EssentCell}$, on several data sets and discuss the results found.</p>","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":"1827-1831"},"PeriodicalIF":0.0,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147849403","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":"Hierarchical Molecular Attention Network: Improving Molecular Property Prediction Through Substructure Identification.","authors":"Liangzhe Chen, Wei Li, Xiaohui Cui, Guoqiang Zhang, Qinghua Li, Haojun Zhu, Jia Wang, Yizhang Jiang","doi":"10.1109/TCBBIO.2026.3697940","DOIUrl":"10.1109/TCBBIO.2026.3697940","url":null,"abstract":"<p><p>Few-shot molecular property prediction is a persisting challenge in many biology-related tasks, because the same molecule may exhibit different properties (e.g., active or inactive) in different tasks. Existing methods view all atoms as equally important and attend to predict the properties by averaging the features of similar molecules, which ignores key substructures within molecules and leads to poor prediction performance. Since a molecule implicitly includes key substructures, which determines the properties of the molecular, and the atom combination forms key substructures, we focus on the atom combination in this paper. With this, we propose the Hierarchical Molecular Attention Network (HMAN) to predict the molecular properties through combining atoms. First, we utilize the average pooling to extract both the prototype and the molecular features, and use Graph Neural Networks (GNN) to extract the atomic features, then concatenates the prototype and molecular features with all atomic features as input to the self-attention mechanism to calculate the different weights for atoms. Here, we select the top $B$ atoms with the highest attention scores to form the key substructures. Second, we view the formed key substructures as the query vectors and regard the molecular features as the key-value pairs, and then feed them into another self-attention to obtain the scores of key substructures. We still select $k$ key substructures with the highest scores, and weight the sum of a molecular feature and $k$ key substructures to predict the molecular properties. To train the HMAN, we design a new loss function, which includes Binary Cross-Entropy and a weighted negative log-likelihood. The former is to predict the molecular properties, and the last one is to optimize the weight distribution, rendering that the weight distribution matches the predicted probability distribution, with back-propagation. Theoretical analysis proves the convergence of the new loss function, and extensive experimental results demonstrate that HMAN significantly outperforms SOTA baseline models in molecular property prediction tasks.</p>","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":"1761-1774"},"PeriodicalIF":0.0,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148045466","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":"scGDCF: Graphical Deep Clustering With Fused Common Information for Single-Cell RNA-Seq Data.","authors":"Yao Dong, Kunyu Li, Yongfeng Dong, Ziyu Ren, Jiaxue Zhang, Yushan Hu, Xuekui Zhang","doi":"10.1109/TCBBIO.2026.3698076","DOIUrl":"10.1109/TCBBIO.2026.3698076","url":null,"abstract":"<p><p>Unsupervised deep clustering plays a crucial role in analyzing single-cell RNA sequencing data (scRNA-seq) as it helps to identify potential cell types. However, most existing clustering methods face challenges in effectively fusing common information between feature and topological structure information, and they may not perform well on the sparse data, which are common in single-cell analysis. To address these challenges, we propose a novel Graphical Deep Clustering with Fused Common Information method for scRNA-seq data, named scGDCF. This method can accurately segregate different cell types even in large and sparse scRNA-seq datasets. In scGDCF, we first introduce a sparse feature representation method that utilizes an adversarial loss function to address the sparsity problem in scRNA-seq data and improve the performance of the discriminator. Next, we design a mutual information extracting operator to deeply mine and fuse the common information from feature and topological structure data, thereby improving the clustering performance. Furthermore, we incorporate the varying degrees of contribution from different neighbor nodes and information sources. To handle this, we promote a dual adaptive attention mechanism that operates at both global and local levels. Finally, experiments on seven real-world datasets and two simulated datasets show scGDCF outperforms 17 state-of-the-art methods. We further extend the clustering results for visualization, analysis of gene differential expression and enrichment, showing scGDCF provides novel insights into cell developmental lineages and preserved inter-cluster distances.</p>","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":"1775-1787"},"PeriodicalIF":0.0,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148146079","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":"WEmarker: Breast Cancer-Specific Prognostic Analysis With Weighted Multiplex Network Embedding.","authors":"Xingyi Li, Junming Li, Zhelin Zhao, Huihui Kong, Min Li, Xuequn Shang","doi":"10.1109/TCBBIO.2026.3693046","DOIUrl":"10.1109/TCBBIO.2026.3693046","url":null,"abstract":"<p><p>In clinical trials, prognostic biomarkers have become essential for guiding treatment decisions after breast cancer surgery. Network-based methods have gained notable attention to reveal marker genes, but many existing methods only focus on a single network, which inevitably neglects the incompleteness of interaction relationships within the network. Even when based upon the multiplex network, most of methods directly integrate the multiplex network into an aggregated network and do not take into account the inherent noise in the biological networks, which can not preserve the topological structure of each original network very well. In this study, we propose a novel method, WEmarker, for breast cancer-specific prognostic analysis. WEmarker reduces the noise level of biological networks and quantifies the probability of interactions between genes, and represents the nodes in the weighted multiplex network as vectors while efficiently retaining the structure information of these networks for identifying prognostic biomarkers. The results show that WEmarker outperforms comparative methods and the case study also demonstrates that biomarkers identified by WEmarker have reliable biological interpretability for breast cancer prognosis.</p>","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":"1635-1642"},"PeriodicalIF":0.0,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147944666","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":"Refined Identification of miRNA-Disease Associations Based on Knowledge-Awareness Propagation.","authors":"Yuliang Cai, Zhen Tang, Wei Qiao, Guiyuan Jiang, Fei Gao, Qiang He, Jiawei Zhang, Wei Qian","doi":"10.1109/TCBBIO.2026.3694582","DOIUrl":"10.1109/TCBBIO.2026.3694582","url":null,"abstract":"<p><p>MicroRNAs (miRNAs) are small non-coding RNAs orchestrating regulatory networks through sequence-specific target recognition. Understanding miRNA-disease correlations is crucial as high-throughput sequencing data growth outpaces experimental validation, necessitating computational approaches for association discovery. Existing frameworks model miRNA-disease interactions as uniform binary relationships, overlooking semantic diversity in different association mechanisms. We propose BKAMDA (MiRNA-Disease Associations prediction Based on Knowledge-Awareness), a novel knowledge-aware model for predicting miRNA-disease associations. Unlike existing methods learning only from miRNA-disease networks, BKAMDA leverages knowledge graphs to delineate distinct association types. By simulating informational propagation within knowledge graphs across diverse miRNA-disease relationships, the model investigates latent connections across various relationship types. Comparative analysis with competitive baselines using real-world experimentally validated datasets demonstrates excellent performance across multiple metrics. Three disease case studies further confirm model accuracy and effectiveness for precision medicine applications. Our knowledge-aware approach significantly advances miRNA-disease association prediction by capturing semantic diversity in biological interactions.</p>","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":"1643-1655"},"PeriodicalIF":0.0,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147978622","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":"HSIC-H-FLapSVM: A Kernel Entropy Component Analysis and Multiple Kernel Learning-Based Fuzzy Laplacian SVM Model for Identifying Exosomal Proteins.","authors":"Jiajia He, Shaoyou Yu, Bo Liao","doi":"10.1109/TCBBIO.2026.3690509","DOIUrl":"10.1109/TCBBIO.2026.3690509","url":null,"abstract":"<p><p>Exosomal proteins participate in many vital biological processes and have great application in clinical diagnosis and prognosis. However, because of the low sequence similarity within exosomal protein datasets. It becomes increasingly urgent to develop computational methods for accurately identifying proteins secreted by exosomes. Therefore, we propose an algorithmic model called HSIC-H-FLapSVM. We extract six feature types from protein sequences. Through experimentation, we determine to use PSSM-DWT, PSSM-AB, and PsePSSM. Subsequently, the features are integrated via multiple kernel learning method based on the Hilbert-Schmidt independence criterion (MKL-HSIC). Following this, fuzzy membership scores for the training samples are derived using kernel entropy component analysis (KECA). Ultimately, HSIC-H-FLapSVM achieves superior performance on testing set, outperforming all competing methods with an ACC of 0.8623 and an MCC of 0.6347. These results demonstrate that HSIC-H-FLapSVM is an effective tool for predicting exosomal proteins.</p>","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":"1585-1595"},"PeriodicalIF":0.0,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147849404","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":"Mamba-ACP: A Hybrid State-Space and Transformer Framework for Interpretable Anticancer Peptide Prediction.","authors":"Adeel Ashraf, Yu-Shan Pang, Abdul Jabbar, Dong-Jun Yu","doi":"10.1109/TCBBIO.2026.3684898","DOIUrl":"10.1109/TCBBIO.2026.3684898","url":null,"abstract":"<p><p>Anticancer peptides (ACPs) represent a promising class of therapeutic agents that selectively destroy cancer cells while sparing healthy tissues. Despite their potential, biological challenges including poor biochemical stability, limited tumor selectivity, and inefficient delivery mechanisms hinder their clinical translation. In parallel, the rapid expansion of peptide sequence data underscores the urgent need for accurate, scalable, and generalizable ACP prediction models. To address these limitations, we propose a robust hybrid deep learning framework, termed as Mamba-ACP-that integrates transformer-based Evolutionary Scale Modeling (ESM-2) embeddings, handcrafted features (AAindex, BLOSUM62), and a Mamba-based sequence modeling architecture. This approach captures both evolutionary and physicochemical properties of peptides to enhance prediction performance. The model was trained and validated by using two benchmark datasets Set 1 and Set 2, commonly used in peptide-based computational biology. Mamba-ACP achieves 87.59% accuracy and an AUC of 0.9356 on Set 1, and 96.69% accuracy and an AUC of 0.9922 on Set 2, beating state-of-the-art ACP predictors like ACP-CapsPred and GRDF by a significant margin. The Mamba-ACP framework processes a token-level fused representation obtained by concatenating the token-level ESM-2 embeddings with the PCA-reduced handcrafted residue descriptors at each sequence position. These results affirm the effectiveness of combining pre-trained transformer embeddings with handcrafted features and structured sequence modeling in improving ACP classification. Our findings position Mamba-ACP as a new benchmark in computational peptide discovery, offering strong generalizability, reduced false positives, and efficient performance. We further provide model-level explanations via gradient-based residue/token saliency, SHAP feature importance for AAindex/BLOSUM62 descriptors, motif-level enrichment, and saliency-guided residue mutation validation.</p>","PeriodicalId":520987,"journal":{"name":"IEEE transactions on computational biology and bioinformatics","volume":"PP ","pages":"1487-1500"},"PeriodicalIF":0.0,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147701888","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}