Bioinformatics (Oxford, England)最新文献

{"title":"RAGCell: retrieval-augmented generation as supervision for versatile single-cell analysis.","authors":"Tianyu Liu, Fan Zhang, Jiayuan Chen, Kun Wang, Haoxuan Li, Shengju Qian, Zhihong Zhu, Donghao Zhou, Hao Wu, Ziheng Zhang, Zhenxi Lin, Xian Wu, Yefeng Zheng","doi":"10.1093/bioinformatics/btag732","DOIUrl":"10.1093/bioinformatics/btag732","url":null,"abstract":"<p><strong>Motivation: </strong>Single-cell foundation models (scFMs) are transforming computational biology by enabling generalizable, task-agnostic representations for versatile single-cell analysis. Despite their progress in facilitating rapid deployment for downstream tasks, off-the-shelf scFMs still have some overlooked concerns: (I) (Pretraining Cost.) Pretrain-based scFMs necessitate pretraining on a vast volume of cells, rendering it draining resources in applications. (II) (Heterogeneous Gap.) Large language model (LLM)-based scFMs ignore the tremendous heterogeneous gap between LLM textual and raw cellular spaces, leading to insufficient capability when facing downstream tasks.</p><p><strong>Results: </strong>To this end, we introduce RAGCell, a versatile single-cell analysis framework that achieves a double-win in both cost-effectiveness and high performance. The success of RAGCell lies in two key aspects: Leveraging LLMs to construct cell-level and feature-level knowledge databases, which serve as supervision signals for training the cell model and significantly reduce the training cost (>pretrain-based scFMs). Aligning cell representations with text embeddings from the bi-level knowledge databases, enabling knowledge transfer from textual spaces to cellular spaces and effectively mitigating the heterogeneous gap (>LLM-based scFMs). Through extensive experiments on five downstream single-cell analysis tasks, we demonstrate that RAGCell achieves outstanding performance compared to state-of-the-art scFMs while operating at less than ∼1/10 the cost of pretrain-based scFMs.</p><p><strong>Availability and implementation: </strong>The source code is available at our supplementary file. Embeddings can be found in https://sites.google.com/yale.edu/scelmolib/home.</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/PMC13675104/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149120184","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":"PhyloZoo: a unified framework for phylogenetic network analysis in Python.","authors":"Niels Holtgrefe","doi":"10.1093/bioinformatics/btag748","DOIUrl":"10.1093/bioinformatics/btag748","url":null,"abstract":"<p><strong>Summary: </strong>Reticulate evolutionary processes (events in which lineages merge, such as hybridization, recombination, and horizontal gene transfer) are widespread across nature but cannot be represented by phylogenetic trees alone. Phylogenetic networks have therefore become an important modelling tool, yet existing software is typically tied to specific inference paradigms and provides limited support for working with multiple network representations in a unified and programmable environment. PhyloZoo is an open-source Python framework that lowers the barrier to developing practical, easy-to-use software for phylogenetic network analysis. It provides data structures and algorithms covering the main representations used in the field, together with dedicated visualization tools and robust I/O for all major phylogenetic file formats. A particular emphasis lies on semi-directed phylogenetic networks, which explicitly represent root uncertainty and have so far received limited support in existing software. By offering a shared, scalable foundation for developing interoperable tools and a combinatorial layer that supports computational proofs and theoretical exploration, PhyloZoo enables reproducible workflows for applied, methodological, and theoretical studies of reticulate evolution.</p><p><strong>Availability and implementation: </strong>PhyloZoo is implemented in Python and installable from PyPI, with source code, documentation, and examples available at https://github.com/nholtgrefe/phylozoo.</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/PMC13676382/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149139643","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":"FAIDR-Desi: interactive software for feature analysis and design of intrinsically disordered regions.","authors":"Tian Hao Huang, Robert Strome, Julie D Forman-Kay, Alan M Moses","doi":"10.1093/bioinformatics/btag715","DOIUrl":"10.1093/bioinformatics/btag715","url":null,"abstract":"<p><strong>Summary: </strong>FAIDR-Desi is a non-technical user interface for design of intrinsically disordered regions (IDRs) based on greedy optimization of sequence features (simple sequence statistics correlated with IDR function, including amino acid composition, residue patterning and short linear interaction motifs), using more features (127) than other IDR design approaches. FAIDR-Desi can be used to design synthetic IDR mimics to establish feature sufficiency for function, or to knock out features to show necessity for function, exemplified with experiments on the yeast Cox15 mitochondrial targeting signal.</p><p><strong>Availability and implementation: </strong>Software is freely available on the web at https://moses-lab.github.io/faidr_desi/ and an archival version of the source code is available on Zenodo (https://doi.org/10.5281/zenodo.22036189).</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/PMC13654200/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149042911","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":"Correction: From atoms to fragments: a coarse representation for efficient and functional protein design.","authors":"","doi":"10.1093/bioinformatics/btag734","DOIUrl":"10.1093/bioinformatics/btag734","url":null,"abstract":"","PeriodicalId":93899,"journal":{"name":"Bioinformatics (Oxford, England)","volume":"42 10","pages":""},"PeriodicalIF":5.4,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13651502/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149148818","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":"Inferring gene expression profiles of tumor clones.","authors":"Shadi Shafighi, Barbara Jurzysta, Ewa Szczurek","doi":"10.1093/bioinformatics/btag717","DOIUrl":"10.1093/bioinformatics/btag717","url":null,"abstract":"<p><strong>Motivation: </strong>Tumors are heterogeneous mixtures of clonal subpopulations of cells that co-evolve in time and space. Apart from genetic differences, cancer clones may reside in different locations in the tumor tissue and show various transcriptional phenotypes, with specific gene expression profiles and marker genes. Computational tools that simultaneously infer genotypes and transcriptional phenotypes of clones in tumor tissues are lacking.</p><p><strong>Results: </strong>We introduce ClonalGE, a probabilistic graphical model identifying cancer clone genotypes, their gene expression profiles, and spatial mapping within tumor tissue. The model proceeds by probabilistic matching of single nucleotide variant sites in spatial transcriptomics to clone-specific variants inferred from DNA sequencing data, and by deconvoluting gene expression in spatial transcriptomics spots into contributions from individual clones. We propose a procedure that utilizes the Monte Carlo posterior samples for identification of differentially expressed genes between clones, using a Highest Density Interval plus Region of Practical Equivalence (HDI+ROPE) criterion that jointly requires credible and practically meaningful effect sizes. ClonalGE demonstrates superior performance on simulated data. Applied to a prostate cancer sample, it reveals clone-specific expression patterns and outperforms its predecessor model in accurately inferring gene expression profiles and explaining spatial transcriptomics signals. Our approach advances the functional and phenotypic characterization of clonal cell subpopulations in tumors.</p><p><strong>Availability: </strong>The implementation of ClonalGE and the data used in this study can be accessed in the following GitHub repository https://github.com/szczurek-lab/ClonalGE/. The benchmarking data alongside the intermediate results are available on Zenodo at DOI: 10.5281/zenodo.20446991.</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/PMC13681668/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149120200","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":"PRISM-GEP: Viewing single-cell expression through the lens of topic modeling.","authors":"Yanir Buznah, Tal Ishon, Uri Shaham","doi":"10.1093/bioinformatics/btag728","DOIUrl":"10.1093/bioinformatics/btag728","url":null,"abstract":"<p><strong>Motivation: </strong>Gene-expression programs (GEPs) are the co-regulated gene modules whose coordinated activity encodes biological processes. Recovering them from single-cell RNA-seq is hard because regulation is many-to-many. A cell can run several programs at once, and a gene can take part in several. Methods that resolve programs along a trajectory break this structure by assigning each gene to a single program.</p><p><strong>Results: </strong>PRISM-GEP discovers overlapping GEPs with Latent Dirichlet Allocation, giving it a prior estimated from gene co-expression in place of the flat default it normally assumes. No reference dataset or pathway database is required. Across 15 human and mouse datasets, PRISM-GEP matches specialized factorizations such as cNMF and scHPF on Gene Ontology Biological Process metrics. Over the 43 rankable dataset-by-metric entries, its mean rank is level with the best, and unlike every specialized method it never ranks last. The same co-expression geometry orders the genes within each program, recovering developmental cascades on par with GeneTrajectory.</p><p><strong>Availability and implementation: </strong>The implementation of our methods, including all code, datasets, and experimental workflows, is available in Python at https://github.com/shaham-lab/PRISM-GEP. The version used in this paper is archived at https://doi.org/10.5281/zenodo.22659056.</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/PMC13661266/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149083224","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":"AcrSeek: metric learning with hybrid negative mining for anti-CRISPR protein detection under extreme class imbalance.","authors":"Chan-Seok Jeong","doi":"10.1093/bioinformatics/btag729","DOIUrl":"10.1093/bioinformatics/btag729","url":null,"abstract":"<p><strong>Motivation: </strong>Anti-CRISPR (Acr) proteins inhibit CRISPR-Cas immunity and are key targets for precise control of CRISPR-based genome editing and phage-host coevolution research. Their experimental identification is costly and low-throughput, so computational predictors are used to prioritise candidates for testing. In this setting validated positives number in the hundreds while phage-derived putative negatives reach tens of thousands, producing extreme class imbalance. Yet existing classification-based predictors train and evaluate on 1:1 balanced datasets and allow high-similarity sequences between training and test, overestimating real-world performance.</p><p><strong>Results: </strong>We present AcrSeek, a metric-learning framework that trains a projection head over a frozen protein language model (PLM) encoder with hybrid negative mining (offline global plus online semi-hard) under a joint triplet-focal objective. Under nested 5-fold cross-validation with leakage controlled by 40% identity clustering, AcrSeek improved AUPRC on the imbalanced (1:383) test set from 0.017 to 0.355 over AcrNET (matched PLM backbone), reaching 0.508 with ESM-2 3B. AcrSeek therefore remains structurally robust under extreme class imbalance and may also serve other protein-function detection problems with scarce positives.</p><p><strong>Availability and implementation: </strong>Source code at https://github.com/jeongchans/acrseek, archived at Zenodo (DOI: 10.5281/zenodo.21946803); data and model checkpoints at Zenodo (DOI: 10.5281/zenodo.20115442).</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/PMC13657427/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149083136","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":"CryoTriplane: triplane-based neural radiance field representation for high-resolution cryo-EM 3D homogeneous reconstruction.","authors":"Yiran Cheng, Bintao He, Qiong Zeng, Fa Zhang, Renmin Han","doi":"10.1093/bioinformatics/btag754","DOIUrl":"10.1093/bioinformatics/btag754","url":null,"abstract":"<p><strong>Motivation: </strong>Determining the three-dimensional (3D) structure of proteins is a cornerstone of structural biology, providing essential insights into their underlying biological mechanisms and functional roles. Cryogenic electron microscopy (cryo-EM) single-particle analysis allows researchers to investigate biomolecules at near-atomic resolutions, unraveling the structure of biomolecular complexes. However, reconstructing high-fidelity 3D volumes from extremely noisy projections, particularly under the interference of structural heterogeneity, remains a challenging task. Traditional Fourier-domain methods often suffer from frequency-dependent signal attenuation and uneven information distribution. While recent deep-learning-based frameworks attempt to predict high-frequency components directly in frequency space, they are highly susceptible to noise and often introduce artifacts. State-of-the-art (SOTA) real-space method circumvents Fourier-domain defects, yet remains constrained by the inherent spectral bias of multi-layer perceptrons, as fixed positional encoding fails to sufficiently capture intricate high-frequency details.</p><p><strong>Results: </strong>We propose cryoTriplane, a real-space 3D homogeneous reconstruction network that utilizes a hybrid spatial representation to adaptively capture fine structural details. By combining fixed positional encoding for noise-robust structural initialization with structure-aware positional encoding for high-frequency refinement, our method achieves 3D reconstructions that are simultaneously globally consistent and locally fine-grained. Comprehensive qualitative and quantitative evaluations across five simulated and four real-world datasets demonstrate that cryoTriplane significantly outperforms existing methods in resolving fine structural features. Furthermore, validation through downstream atomic modeling underscores cryoTriplane's ability to recover high-quality density maps. Specifically, our method achieves sequence coverage improvements of 4.6% and 5.5% on the EMPIAR-10005 and EMPIAR-10492 datasets over the SOTA approach, respectively.</p><p><strong>Availability and implementation: </strong>Our code is available at https://github.com/chengyr1999/CryoTriplane. The real datasets used in this study are publicly available from EMPIAR.</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/PMC13690421/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149186826","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":"CAD-score-LT: versatile contact area-based comparison of macromolecular structures, interfaces, and binding sites.","authors":"Kliment Olechnovič","doi":"10.1093/bioinformatics/btag756","DOIUrl":"10.1093/bioinformatics/btag756","url":null,"abstract":"<p><strong>Motivation: </strong>Comparison of macromolecular structural models requires similarity measures that remain informative for complexes, interfaces, and binding sites. Widely used scores emphasize global superposition or distance deviations and are not designed primarily for direct assessment of contact patterns. A fast and flexible contact-area-based alternative is therefore highly relevant.</p><p><strong>Results: </strong>We present CAD-score-LT, a high-performance implementation of the Contact Area Difference score (CAD-score) for superposition-free comparison of macromolecular structures. CAD-score-LT evaluates tessellation-derived contact, binding-site, and solvent-accessible surface areas at atom, residue, and chain levels, and supports selective scoring of substructures, interfaces, and binding regions. The software also provides sequence-guided preprocessing, chain remapping, symmetry-adjusted ligand scoring, local scores output, and clustering. Benchmarking against previous CAD-score implementations on CASP16 oligomer model pairs showed near-identical scores but more than one order of magnitude lower running times, with speedups reaching up to 100-fold for whole-structure scoring and 200-fold for inter-chain interface scoring. Case studies in supplementary material demonstrate utility for large-scale model clustering, analysis of protein-nucleic acid interfaces, and evaluation of protein-ligand docking models.</p><p><strong>Availability and implementation: </strong>A repository with the CAD-score-LT software and the reproducible benchmarking and example workflows is available at https://github.com/kliment-olechnovic/cadscore-lt-supplementary-info and archived at Zenodo https://doi.org/10.5281/zenodo.21983528. The latest CAD-score-LT software is openly available under MIT license as a standalone C++ application at https://www.voronota.com/expansion_lt_cadscore/, and as a Python library at https://pypi.org/project/cadscorelt/.</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/PMC13686761/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149169913","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":"FMCerNet: frequency-aware multi-label learning for fine-grained cervical cytology screening.","authors":"Linyun Zhou, Qiuyu Cai, Xiaoling Wang, Weihan Li, Jian Yang, Mingli Song, Jing Zhang, Feng Zhou, Qihan You, Yudong Wang, Zunlei Feng, Xiuming Zhang","doi":"10.1093/bioinformatics/btag759","DOIUrl":"10.1093/bioinformatics/btag759","url":null,"abstract":"<p><strong>Motivation: </strong>Automated cytology screening is essential for early cervical cancer prevention, where gigapixel whole-slide images (WSIs) are typically partitioned into tiles for tile-level feature extraction and subsequent WSI-level diagnosis aggregation. However, existing tile-level methods remain constrained by limited input sizes, which frequently obscures sparse and morphologically subtle abnormal cells. Furthermore, these methods commonly formulate tile-level representation learning as a coarse single-label classification task, thereby failing to capture the complex co-occurrence patterns and morphological heterogeneity of cervical cytological abnormalities.</p><p><strong>Results: </strong>We propose the Frequency-Aware Multi-Label Learning Network (FMCerNet), a tile-level feature extractor equipped with a Large-Field Frequency-Enhanced ViT backbone and a Coarse-to-Fine multi-label classification head. The former incorporates a Dual-Tree Complex Wavelet Transform module to efficiently extract frequency-domain features from large-scale tiles, whereas the latter enhances fine-grained tile representation decoding through hierarchical supervision. Experiments on both public and private datasets demonstrate that FMCerNet improves the accuracy of positive screening and fine-grained subtype classification while maintaining high inference speed.</p><p><strong>Availability and implementation: </strong>Code at https://github.com/imAzhou/FMCerNet and archived on Zenodo (doi: 10.5281/zenodo.22886105).</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":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149221071","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}