Bioinformatics (Oxford, England)最新文献

{"title":"ShadoNet: a cell detection and classification framework for Ki-67 pathology images.","authors":"Mahsa Ghasemi, Fuyong Xing, Toby C Cornish, Debashis Ghosh, Jiang Bian, Xuhong Zhang","doi":"10.1093/bioinformatics/btag704","DOIUrl":"10.1093/bioinformatics/btag704","url":null,"abstract":"<p><strong>Motivation: </strong>Accurate detection and classification of individual cells are essential for quantitative histopathology. In Ki-67-based tumor grading, the clinically relevant index is calculated from immunopositive and immunonegative tumor cells, while non-tumor cells should be excluded. This is challenging because Ki-67-stained tumor regions often contain a heterogeneous mixture of immunopositive tumor cells, immunonegative tumor cells, and non-tumor cells. Here, we study cell detection and classification in Ki-67-stained histopathology images to support grading of pancreatic neuroendocrine tumors (PanNETs), where reliable tumor-cell-specific identification is critical for tumor grading and decision-making.</p><p><strong>Results: </strong>We present ShadoNet, a U-Net-style encoder-decoder framework that integrates shape priors with structured regression for cell detection and classification. ShadoNet predicts class-specific proximity maps that emphasize cell centers and jointly encode spatial location, morphology, and class identity through smooth intensity decay from cell centers. Across multiple Ki-67-stained histopathology datasets, incorporating shape information improved performance on standard detection and classification metrics, indicating that shape-derived priors provide useful class-related cues-without requiring full instance-level boundary annotations.</p><p><strong>Availability and implementation: </strong>ShadoNet uses the Segment Anything Model (SAM) to generate initial cellular shape masks, which are refined using human-annotated nuclear point labels to construct ground-truth proximity maps. During training, shape guidance is enforced via two auxiliary objectives: (i) a rotation-extended Scaled IoU (SIoU) loss to capture orientation and shape alignment, and (ii) a Hausdorff Distance Transform (HDT) loss to penalize boundary discrepancies between predictions and ground truth. Code is publicly available at: https://github.com/GhasemiGOF/ShadoNet. A fixed version of the code used for the experiments in this manuscript has been archived at Zenodo with DOI: https://doi.org/10.5281/zenodo.21050210. BCData and SHIDC-B-Ki-67 are publicly available datasets and can be accessed through their original publications.</p>","PeriodicalId":93899,"journal":{"name":"Bioinformatics (Oxford, England)","volume":" ","pages":""},"PeriodicalIF":5.4,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13653216/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149070678","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Pocket contrastive learning and conformation fusion for robust structure-conditioned 3D molecular generation under receptor conformational variation.","authors":"Huan Liu, Yijie Ding, Xiaoyi Guo, Hongjie Wu","doi":"10.1093/bioinformatics/btag740","DOIUrl":"10.1093/bioinformatics/btag740","url":null,"abstract":"<p><strong>Motivation: </strong>Pocket-conditioned molecular generation incorporates the three-dimensional (3D) geometry and chemical environment of protein binding sites into molecular design. However, receptor conformational variation can introduce distribution shifts between training and inference, particularly under cross-docking conditions. Existing approaches often do not fully exploit the correspondence between related receptor conformations during training.</p><p><strong>Results: </strong>We propose a structure-conditioned 3D molecular generation framework that combines pocket-level contrastive learning with reference-conformation feature fusion. Built on a stepwise constructive generative model, these training-stage components encourage stable pocket representations across related conformations while preserving single-pocket inference without requiring a reference conformation. Evaluation on CrossDocked2020 and three external datasets, ASB-E, DUD-E, and APObind, demonstrates improvements in generation success, structural validity, and docking-related performance, particularly under cross-dock evaluation. These results support improved robustness of pocket-conditioned molecular generation under receptor conformational variation.</p><p><strong>Availability and implementation: </strong>The source code and reproduction instructions are available at https://github.com/LHHHHuan/MTMGen. Pretrained models and an archival code snapshot are available at https://doi.org/10.5281/zenodo.21944857.</p>","PeriodicalId":93899,"journal":{"name":"Bioinformatics (Oxford, England)","volume":" ","pages":""},"PeriodicalIF":5.4,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13671754/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149120129","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"OrchAlign: orchestrated multimodal alignment for gene expression prediction.","authors":"Li Liu, Guipeng Xv, Shujie Liu, Yeyun Gong, Chen Lin","doi":"10.1093/bioinformatics/btag711","DOIUrl":"10.1093/bioinformatics/btag711","url":null,"abstract":"<p><strong>Motivation: </strong>Gene expression prediction benefits from integrating DNA sequences and epigenomic signals. Existing approaches typically combine these modalities using simple operations such as concatenation or summation, without explicitly modeling fine-grained token-level cross-modal interactions. Establishing precise cross-modal correspondences remains challenging due to (1) inter-modal discrepancies, and (2) intra-modal structural preservation.</p><p><strong>Results: </strong>To address these challenges, we propose OrchAlign, an orchestrated multimodal alignment framework that establishes fine-grained cross-modal alignment while preserving intra-modal structure. Our approach first disentangles each modality into shared and specific representations. We then perform focused cross-modal alignment on the shared subspace within a key regulatory region, producing aligned representations with token-level correspondences. Moreover, we enforce consistency between relational structures of the aligned shared representations and their corresponding modality-specific counterparts, enhancing intra-modal structural coherence. Finally, the aligned shared representations and modality-specific representations are passed to the BiMamba backbone for further multimodal fusion. Our experimental results show that OrchAlign achieves state-of-the-art performance in gene expression prediction.</p><p><strong>Availability and implementation: </strong>All resources are available at https://github.com/XMUDM/OrchAlign.</p>","PeriodicalId":93899,"journal":{"name":"Bioinformatics (Oxford, England)","volume":" ","pages":""},"PeriodicalIF":5.4,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13661263/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149083226","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"VCCV: conservative transcriptomic corroboration for measurement prioritization of computational drug-target hypotheses.","authors":"Haihui Huang, Yanan Zhou, Dingkui Kang, Yong Liang","doi":"10.1093/bioinformatics/btag694","DOIUrl":"10.1093/bioinformatics/btag694","url":null,"abstract":"<p><strong>Motivation: </strong>Computational drug-target interaction (DTI) models nominate plausible binders but cannot determine which candidate best accounts for an observed cellular response. Perturbational transcriptomics offers orthogonal mechanistic evidence, yet pharmacology-to-genetics mismatch, incomplete reference coverage, and non-specific stress programs make simple signature matching unreliable. This motivates a principled integration layer that corroborates hypotheses conservatively, abstains under global mismatch, and prioritizes informative follow-up measurements when the evidence remains ambiguous.</p><p><strong>Results: </strong>We present Virtual-to-Cellular Corroboration for Validation (VCCV), a model-agnostic posterior-triage layer for pre-trained DTI models. VCCV updates calibrated DTI working weights with context-aligned perturbational evidence using a near-identity affine map and exact covariance transport. Within the stated Gaussian class, exact transport is the unique uncertainty update that preserves posterior-odds comparisons under invertible affine changes of measurement coordinates. VCCV also introduces an empirical warning branch for abstention and converts residual ambiguity into compact follow-up gene panels, using a submodular objective for deep near-ties. Each query is assigned one of three actionable states: a target-resolved hypothesis, an abstention, or a prioritized panel. Across five DTI models, VCCV improved discrimination (paired ROC-AUC gains 0.021-0.037) and reduced negative log-likelihood in every case. Additional evaluations showed improved ranking on the same-cell-supported endpoint, warning-score discrimination of strong-response profiles (ROC-AUC 0.876), and better recovery of full-coordinate leading hypotheses by selected panels than by random panels. Across these retrospective kinase-focused evaluations, VCCV provides a principled bridge from computational nomination to conservative, measurement-directed cellular corroboration.</p><p><strong>Availability and implementation: </strong>Source code of VCCV is publicly available at https://github.com/bio-ai-source/VCCV.</p>","PeriodicalId":93899,"journal":{"name":"Bioinformatics (Oxford, England)","volume":" ","pages":""},"PeriodicalIF":5.4,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13633670/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149003146","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Mechanistic interpretability of fine-tuned protein language models for nanobody thermostability prediction.","authors":"Taihei Murakami, Yuki Hashidate, Yasuhiro Matsunaga","doi":"10.1093/bioinformatics/btag685","DOIUrl":"10.1093/bioinformatics/btag685","url":null,"abstract":"<p><strong>Motivation: </strong>While Protein Language Models (PLMs) fine-tuned on biophysical data achieve high predictive accuracy, the physical principles underlying their predictions remain obscure. Deciphering these representations offers a unique opportunity to not only interpret model decisions but also to discover novel biophysical insights governing protein properties. Here, we present a framework using Sparse Autoencoders (SAEs) to extract mechanistic knowledge from PLMs fine-tuned for nanobody thermostability.</p><p><strong>Results: </strong>We fine-tuned the ESM-2 model on the nanobody thermostability dataset, achieving superior performance compared to significantly larger state-of-the-art models. SAE analysis successfully decomposed the model's dense embeddings into sparse, interpretable features without loss of predictive accuracy. We characterized these features through both global and local analyses. Global analysis provided an aggregate map of position-dependent feature contributions, whereas local analysis identified specific residue-level patterns, including known determinants such as the VHH-tetrad and critical disulfide bonds, as well as candidate stabilizing residues. Free Energy Perturbation calculations supported the structural plausibility of selected residue-level hypotheses. These results show that SAE-based interpretation can generate testable, structurally grounded hypotheses for rational protein engineering.</p><p><strong>Availability: </strong>The data and source code of the proposed method are available at GitHub (https://github.com/matsunagalab/paper_nanobody-thermostability-sae) and Zenodo (DOI: 10.5281/zenodo.18012027).</p>","PeriodicalId":93899,"journal":{"name":"Bioinformatics (Oxford, England)","volume":" ","pages":""},"PeriodicalIF":5.4,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13637583/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149013361","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ProDTI: prior-guided positive-unlabeled learning for robust drug-target interaction prediction.","authors":"Yu Wang, Haoxiang Zhang, Ziye Zhu, Qiang Wei, Yun Li","doi":"10.1093/bioinformatics/btag712","DOIUrl":"10.1093/bioinformatics/btag712","url":null,"abstract":"<p><strong>Motivation: </strong>Accurate Drug-Target Interaction (DTI) prediction is critical for drug discovery. While deep learning advances this field, the fundamental lack of verified negative interactions inevitably causes biased risk estimation. Although formulating DTI prediction as a Positive-Unlabeled (PU) learning problem is theoretically sound, directly applying PU learning to deep models suffers from early-stage optimization instability and decision boundary ambiguity.</p><p><strong>Results: </strong>We propose ProDTI, a prior-guided PU learning framework integrating domain knowledge into the PU objective. We construct a hybrid prior combining explicit biochemical similarities with implicit semantic representations from large-scale pre-trained models. This prior drives two synergistic mechanisms: soft-label distillation to stabilize the optimization trajectory, and hard-anchor regularization to explicitly penalize reliable negatives and sharpen decision boundaries. Evaluated on five benchmarks, ProDTI consistently outperforms state-of-the-art baselines, exhibiting superior robustness in challenging cold-start scenarios and providing a reliable tool for computational drug discovery.</p><p><strong>Availability and implementation: </strong>Source code and data are available at https://github.com/WangYu-AI4SCI/ProDTI.</p>","PeriodicalId":93899,"journal":{"name":"Bioinformatics (Oxford, England)","volume":" ","pages":""},"PeriodicalIF":5.4,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13676319/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149129391","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"HINN prioritizes multi-omics biomarkers associated with cognitive decline.","authors":"Yashu Vashishath, Sarah Anne Beaver, Fahad Saeed, Serdar Bozdag","doi":"10.1093/bioinformatics/btag741","DOIUrl":"10.1093/bioinformatics/btag741","url":null,"abstract":"<p><strong>Motivation: </strong>Integrating multi-omics data for disease prediction remains challenging due to the complexity of cross-layer biological interactions and the limited use of prior biological knowledge in many machine learning models. We present the Hierarchical Input Neural Network (HINN), a deep learning framework that incorporates biologically informed connections across genomic, epigenetic, and transcriptomic layers to model regulatory relationships in a structured and interpretable manner. HINN was applied to blood-derived multi-omics data to predict cognitive assessment scores, including Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment, Alzheimer's Disease Assessment Scale-Cognitive Subscale 11 tasks (ADAS11), and Rey Auditory Verbal Learning Test. A biologically guided feature construction pipeline was used to integrate genome-wide association studies-derived single nucleotide polymorphisms (SNPs), promoter-region DNA methylation, and gene expression data through Gene Ontology Biological Process and KEGG pathway mappings.</p><p><strong>Results: </strong>The model was evaluated using a rigorous training strategy that separates feature selection from testing and incorporates cross-validation as well as repeated runs to ensure robustness. Across all cognitive outcomes, HINN achieved the lowest mean normalized mean squared error among the evaluated models, with statistically significant improvements over most baseline methods. Ablation analyses showed that the contributions of individual omics modalities and biologically informed connectivity varied across cognitive outcomes, with the clearest benefits of structured connectivity observed for ADAS11 and MMSE. Feature attribution analysis identified a multi-omics cascade involving SNP (rs116557230), CpG site (cg24041822), and a SOCS4 gene expression probe, which showed consistent associations with multiple cognitive outcomes. These findings highlight the ability of HINN to uncover biologically meaningful cross-omics relationships relevant to cognitive decline. Together, this work demonstrates that integrating deep learning with biological knowledge enables biomarker discovery for complex diseases.</p><p><strong>Availability and implementation: </strong>All code, sample data, and instructions are publicly available via GitHub (https://github.com/bozdaglab/HINN) and archived on Zenodo (https://doi.org/10.5281/zenodo.19197057) to ensure reproducibility and long-term accessibility.</p>","PeriodicalId":93899,"journal":{"name":"Bioinformatics (Oxford, England)","volume":" ","pages":""},"PeriodicalIF":5.4,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13684270/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149129313","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integration of polygenic risk with single cell methylation data using met-scDRS.","authors":"Xinzhe Li, Zhuozheng Shi, Kangcheng Hou, Katherine W Eyring, Cuining Liu, Matthew G Heffel, Chongyuan Luo, Daniel H Geschwind, Bogdan Pasaniuc","doi":"10.1093/bioinformatics/btag751","DOIUrl":"10.1093/bioinformatics/btag751","url":null,"abstract":"<p><strong>Motivation: </strong>Although large scale genome-wide association studies (GWAS) have identified many risk variants for psychiatric traits their functional implication remain largely unknown. One potential mechanism for their risk is through cellular context-dependent epigenetics such as methylation. Previous methodologies have focused on integrating GWAS with single cell transcriptomics, which allowed identification of within-cell-type heterogeneity in disease risk association. We extended this framework to provide a method called met-scDRS that integrates GWAS with single cell epigenomics and investigates how psychiatric disease risk can manifest through methylomics at single cell resolution.</p><p><strong>Results: </strong>We propose an approach that integrates single cell methylomes with disease associated genes identified from GWAS at single-cell resolution. We found our method to be well calibrated and powered in different scenarios. We then analysed a publicly available single cell methylome dataset of the human brain for 75 different traits and generated disease relevance scores for each cell for each trait. We find high brain specificity with major cell type enrichments congruent with known findings. We identify within-cell-type heterogeneity for psychiatric disease risk along the brain spatial axis as well as subpopulations that are differentially associated with diseases. Finally, we showed that genes with high correlation between methylation levels and disease relevance scores to be highly enriched in specific functional pathways such as synaptic formation and transmission.</p><p><strong>Availability: </strong>met-scDRS is freely available in GitHub as a python package under MIT license: https://github.com/bogdanlab/met-scDRS. Source code is also available from Zenodo with DOI: 10.5281/zenodo.21841506. All dataset used in this work are publicly available.</p>","PeriodicalId":93899,"journal":{"name":"Bioinformatics (Oxford, England)","volume":" ","pages":""},"PeriodicalIF":5.4,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13693840/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149229188","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"METANet: a supervised ensemble learning framework for reconstructing direct and functional tissue-specific transcription factor networks.","authors":"Wooseok J Jung, Sandeep Acharya, Daniel P Ruskin, Shu Liao, Vaha Akbary Moghaddam, Zolboo Erdenebaatar, Michael R Brent","doi":"10.1093/bioinformatics/btag738","DOIUrl":"10.1093/bioinformatics/btag738","url":null,"abstract":"<p><strong>Motivation: </strong>Reconstructing tissue-specific transcription factor (TF) networks remains challenging. TF-motif-based methods often lack functional validation, while expression-based methods struggle to distinguish direct binding from indirect regulation. Integration of diverse data types is necessary to accurately prioritize likely functional targets directly bound by TFs across human tissues.</p><p><strong>Results: </strong>We introduce METANet (Motif Expression TF Association Networks), a supervised ensemble learning framework that combines TF motifs, cis-regulatory element activity, and linear and nonlinear expression-derived features to predict TF binding. Applied to 36 human tissues, predicted METANet maps significantly outperform established methods in identifying direct, likely functional targets of TFs validated by ChIP-seq and gene ontology. Furthermore, METANet maps capture tissue-specific regulation comparable to existing methods, allowing the identification of reproducible gene-trait associations.</p><p><strong>Availability and implementation: </strong>All code and network maps are freely available at Zenodo https://doi.org/10.5281/zenodo.20614479.</p>","PeriodicalId":93899,"journal":{"name":"Bioinformatics (Oxford, England)","volume":" ","pages":""},"PeriodicalIF":5.4,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13693595/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149129389","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Hierarchical breakdown of RNA structure prediction in CASP16: from reliable local helices to speculative multimer assembly.","authors":"Chandran Nithin, Smita P Pilla, Sebastian Kmiecik","doi":"10.1093/bioinformatics/btag689","DOIUrl":"10.1093/bioinformatics/btag689","url":null,"abstract":"<p><strong>Motivation: </strong>CASP16 provided a community-wide benchmark for assessing RNA structure prediction, including the first large-scale blind assessment of RNA-RNA multimer prediction. CASP16 results showed that accurate three-dimensional modeling, especially for RNA-RNA multimers, remains a major challenge across the field.</p><p><strong>Results: </strong>In this work, we use the submissions of our group (LCBio) as a diagnostic case study to examine the current limits of RNA structure prediction. In the official CASP16 best-of-submitted-models analysis, our workflow ranked first in the RNA-RNA multimer category and remained competitive for monomers. This makes the submitted model set useful for examining why high-ranking multimer predictions can still deviate substantially from experimental structures. We combine hierarchical analysis with representative case studies to connect this field-wide limitation to specific structural failure modes, showing that prediction accuracy decreases from relatively reliable canonical base-pairing and local helical organization to less reliable non-canonical interactions, stacking geometry, tertiary motifs, and assembly-level features. In RNA-RNA multimers, errors in monomer structure can combine with uncertainty in interface geometry and model selection, reducing the accuracy of the assembled complexes. These findings point to monomer structure accuracy, interface modeling, and model selection as key areas for improving RNA-RNA multimer prediction.</p><p><strong>Availability and implementation: </strong>The scripts used for feature extraction, scoring, bootstrap confidence-interval estimation, and figure generation are available at Zenodo: https://doi.org/10.5281/zenodo.21393731.</p>","PeriodicalId":93899,"journal":{"name":"Bioinformatics (Oxford, England)","volume":" ","pages":""},"PeriodicalIF":5.4,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13645652/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148979095","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}