Bioinformatics advances最新文献

{"title":"Benchmarking antibody modeling tools across structure prediction, docking, and paratope-epitope interface analysis.","authors":"Zeyuan Yu, Jilei Wu, Ziyao Ning, Chunxia Qiao, Jing Wang, Xinying Li, Chenghua Liu, Guojiang Chen, Jiannan Feng, Jijun Yu","doi":"10.1093/bioadv/vbag223","DOIUrl":"10.1093/bioadv/vbag223","url":null,"abstract":"<p><strong>Motivation: </strong>Computational antibody engineering requires reliable prediction of antibody variable-fragment structures, antigen-antibody complexes, and binding interfaces. However, publicly available tools for these tasks have rarely been compared across the complete workflow under a controlled and statistically grounded design.</p><p><strong>Results: </strong>We evaluated ImmuneBuilder, IgFold, AlphaFold3, GRAMM, and dyMEAN on 50 non-redundant humanized antibody-antigen complexes using multiple retained predictions and paired statistical testing. All three antibody structure predictors were accurate, with AlphaFold3 performing best overall and for the third complementarity-determining region of the heavy chain. AlphaFold3 also substantially outperformed GRAMM and dyMEAN in complex prediction, producing medium- or high-quality binding interfaces for 46% of the complexes, although overall interface accuracy remained limited. When docking was reliable, AlphaFold3 accurately recovered epitope and paratope residues, salt bridges, and non-bonded contacts, but reproduced hydrogen bonds and fine-grained contact strengths less consistently. These findings provide practical guidance for selecting tools across antibody-modeling workflows and identify persistent limitations in fine-grained interface prediction.</p><p><strong>Availability and implementation: </strong>Data, structural predictions, evaluation results, and analysis code are available from Zenodo under record 20710876.</p>","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"6 1","pages":"vbag223"},"PeriodicalIF":2.6,"publicationDate":"2026-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13557113/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148921861","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":"TimeTraits: extracting functional traits from biological time-series in R.","authors":"Sarah C L Lock, Marina I Knight, Seth J Davis, Daphne Ezer","doi":"10.1093/bioadv/vbag226","DOIUrl":"10.1093/bioadv/vbag226","url":null,"abstract":"<p><strong>Motivation: </strong>It is currently challenging for scientists to extract meaningful parameters from circadian rhythms that dynamically change over time (i.e. non-stationary rhythms). Circadian rhythms may be non-stationary due to dynamic conditions, such as changes in photoperiod. Circadian rhythms are also nonstationary when plants are suddenly placed under continuous light conditions after a period of entrainment, as their clocks slowly drift without external entrainment signals.</p><p><strong>Results: </strong>TimeTraits provides a series of functions to enable the extraction of parameters from time series data using Functional Data Analysis methods. We demonstrate its utility in dissecting the changes in curve shape of a circadian clock bioluminescence marker, following a photoperiod shift in wildtype and phyB Arabidopsis.</p><p><strong>Availability and implementation: </strong>TimeTraits is available in CRAN: https://cran.r-project.org/web/packages/TimeTraits.</p>","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"6 1","pages":"vbag226"},"PeriodicalIF":2.6,"publicationDate":"2026-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13537318/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148882038","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":"Towards improved particle reconstruction for single-molecule localization microscopy using geometric deep learning.","authors":"Diana Mindroc-Filimon, Dominic A Helmerich, Patrick Salome, Markus Sauer, Philip Kollmannsberger","doi":"10.1093/bioadv/vbag228","DOIUrl":"10.1093/bioadv/vbag228","url":null,"abstract":"<p><strong>Motivation: </strong>Single-molecule localization microscopy (SMLM) can resolve intracellular structures down to the nanoscale, but often produces sparse and incomplete data. Particle averaging (PA) can aid with the reconstruction of complete structures, but traditional PA methods can suffer from template bias or the high computational costs of geometric alignment.</p><p><strong>Results: </strong>To address these limitations, we developed a geometric deep learning (GDL) framework for enhanced, template-free 3D particle averaging. Our pipeline uses a GDL autoencoder, trained on high-fidelity simulated data, to map incomplete point clouds into a robust latent space. By averaging feature vectors directly within this space, our method bypasses the need for explicit 3D alignment. We validated our approach on simulated DNA origami and experimental nuclear pore complex (NPC) data. The latent space averaging successfully reconstructed NPC structures with key metrics (ring radius <math><mrow><mo>≈</mo> <mn>46</mn> <mtext>nm</mtext></mrow> </math> , ring distance <math><mrow><mo>≈</mo> <mn>52</mn> <mtext>nm</mtext></mrow> </math> ) that are comparable to state-of-the-art methods. This work establishes a viable GDL pipeline for SMLM analysis, offering an efficient alternative to traditional PA. While the current model requires structure-specific training, our results highlight the significant potential of GDL for quantitative structural biology.</p><p><strong>Availability: </strong>https://github.com/dianamindroc/smlm.</p>","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"6 1","pages":"vbag228"},"PeriodicalIF":2.6,"publicationDate":"2026-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13619314/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149076932","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":"tidyexposomics: integrated exposure-omics analysis powered by tidy principles.","authors":"Jason Laird, Thomas Hartung, Fenna C M Sillé, Alexandra Maertens","doi":"10.1093/bioadv/vbag220","DOIUrl":"10.1093/bioadv/vbag220","url":null,"abstract":"<p><strong>Motivation: </strong>Environmental exposures shape health across the life course, influencing molecular pathways, disease susceptibility, and therapeutic response. However, integrating exposome and multi-omics data poses several challenges, ranging from extensive analytical steps to ensuring trends are consistent across studies.</p><p><strong>Results: </strong>To address this challenge, we developed tidyexposomics, an open-source R package that delivers an end-to-end, tidyverse-native workflow including ontology-based exposure annotation, quality control, association testing, stability assessment, multi-omics integration, network analysis, and functional enrichment. The package offers intuitive Application Programming Interface (API), modular functions, and extensive visualization tools to facilitate flexible analyses. Comprehensive analytical step tracking and result export functions enable reproducibility and consistent reporting of downstream results. By combining systematic preprocessing, statistical modeling and biologically informed interpretation in one package, tidyexposomics streamlines exposure-omics analysis.</p><p><strong>Availability and implementation: </strong>The GitHub repository for tidyexposomics is available at: https://github.com/BioNomad/tidyexposomics.</p><p><strong>Pre-processed example data are available at: </strong>https://doi.org/10.5281/zenodo.17049350.</p>","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"6 1","pages":"vbag220"},"PeriodicalIF":2.6,"publicationDate":"2026-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13532756/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148876672","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":"Usage and impact of global biodata resources.","authors":"Carla Cummins, Heidi Imker, Charles E Cook, Guy Cochrane","doi":"10.1093/bioadv/vbag219","DOIUrl":"10.1093/bioadv/vbag219","url":null,"abstract":"<p><strong>Motivation: </strong>Biodata resources constitute a critical, large-scale, and globally distributed infrastructure underpinning life science research, yet their organic growth has hindered efforts to quantify key indicators needed to justify sustainable support, including usage, impact, and interdependencies. Here, we present an updated Global Biodata Coalition inventory alongside a Total Resource Usage (TRU) dataset that integrates this inventory with two complementary literature-derived sources: data citations and informal resource name mentions extracted from full-text articles using a fine-tuned machine learning model. A unified database schema enables cross-resource comparisons, dependency network analyses, and evaluation of resource name distinctiveness.</p><p><strong>Results: </strong>The combined dataset captures 11.5 million formal and informal references, revealing that most resources are acknowledged informally within article text. Network analysis indicates a densely interconnected ecosystem in which Global Core Biodata Resources function as key providers and integrators, underscoring their foundational role. While full resource names are generally distinctive, widespread use of acronyms limits detectability through text mining. Together, these findings provide robust empirical evidence of a highly utilized and interconnected biodata infrastructure, highlight limitations of single-metric assessments, and underscore the need for multi-dimensional evaluation frameworks and more consistent data citation practices to support informed decision-making and long-term sustainability.</p><p><strong>Availability and implementation: </strong>The database and analytical code described here are available on https://github.com/globalbiodata.</p>","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"6 1","pages":"vbag219"},"PeriodicalIF":2.6,"publicationDate":"2026-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13532761/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148876669","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":"AmpliPhy improves gene trees by adding homologous sequences without affecting alignments.","authors":"Dongwook Kim, Manuel Gil, Kazutaka Katoh, Christophe Dessimoz","doi":"10.1093/bioadv/vbag222","DOIUrl":"10.1093/bioadv/vbag222","url":null,"abstract":"<p><strong>Motivation: </strong>In phylogenomics, gene tree reconstruction depends on multiple sequence alignment and tree inference, and ongoing work continues to improve inference quality. Denser taxon sampling has been associated with improved gene tree inference, suggesting that adding homologs could be a practical route to higher accuracy as sequence databases continue to expand. However, adding sequences can influence multiple steps of typical inference pipelines, and little is known on its specific effect on the multiple sequence alignment, tree reconstruction, and rooting steps.</p><p><strong>Results: </strong>We performed a large-scale empirical and simulated benchmarks to quantify how homolog enrichment affects alignment and phylogenetic inference. Using an enrichment-impoverishment design and a measure of tree accuracy based on taxonomic congruence, we found that enrichment consistently improves tree inference quality, while effects on alignment quality are marginal. We show that this improvement is associated with, but not restricted to accurate root placement on enriched trees when sensitive homolog search is accompanied. Notably, much of the benefit can be retained with relatively compact alignments produced by sequence addition. Building on these observations, we provide a tool, AmpliPhy, which efficiently improves phylogenetic reconstruction of protein families through homolog enrichment.</p><p><strong>Availability and implementation: </strong>The AmpliPhy open-source pipeline software is available at https://github.com/DessimozLab/ampliphy. The scripts used to generate the data and figures are available from https://github.com/DessimozLab/ampliphy-analysis.</p>","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"6 1","pages":"vbag222"},"PeriodicalIF":2.6,"publicationDate":"2026-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13516908/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148835192","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":"LAFA: a framework for reproducible longitudinal assessment of protein function annotation models.","authors":"An Phan, Yanli Wang, Frimpong Boadu, Jianlin Cheng, Predrag Radivojac, Iddo Friedberg","doi":"10.1093/bioadv/vbag221","DOIUrl":"10.1093/bioadv/vbag221","url":null,"abstract":"<p><strong>Motivation: </strong>Protein function prediction is a challenging task and an open problem in computational biology. The Critical Assessment of protein Function Annotation (CAFA) is a triennial, community-driven initiative that provides an independent, large-scale evaluation of computational methods for protein function prediction through time-delayed benchmarking experiments. CAFA has played a key role in highlighting high-performing methodologies and fostering detailed analysis and exchange of ideas. However, outside the periodic CAFA challenges, there is no platform for the continuous evaluation of newly developed methods and tracking performance as function annotations accumulate.</p><p><strong>Results: </strong>Here we introduce the Longitudinal Assessment of Protein Function Annotation Models server (LAFA) as a persistent benchmarking system for protein function prediction methods. LAFA provides a continuous evaluation of containerized function prediction methods, enabling up-to-date and robust comparative assessment of method performance under evolving ground truth. LAFA accelerates methodological iteration, supports reproducibility, and offers a more dynamic and fine-grained view of progress in protein function prediction.</p><p><strong>Code and data availability: </strong>LAFA is available at https://functionbench.net/. Detailed evaluation results can be found at https://github.com/FriedbergLab/CAFA_forever.</p>","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"6 1","pages":"vbag221"},"PeriodicalIF":2.6,"publicationDate":"2026-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13501529/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148815084","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":"seqme: a Python library for evaluating biological sequence design from generative models.","authors":"Rasmus Møller-Larsen, Adam Izdebski, Jan Olszewski, Pankhil Gawade, Michal Kmicikiewicz, Wojciech Zarzecki, Ewa Szczurek","doi":"10.1093/bioadv/vbag212","DOIUrl":"10.1093/bioadv/vbag212","url":null,"abstract":"<p><strong>Summary: </strong>Recent advances in computational methods for designing biological sequences have sparked the development of metrics to evaluate these methods performance in terms of the fidelity of the designed sequences to a target distribution and their attainment of desired properties. However, a software library implementing these metrics was lacking. In this work we introduce seqme, a modular and highly extendable open-source Python library, containing model-agnostic metrics for evaluating computational methods for biological sequence design. seqme considers three groups of metrics: sequence-based, embedding-based, and property-based, and is applicable to a wide range of biological sequences: small molecules, DNA, ncRNA, mRNA, peptides and proteins. The library offers a number of embedding and property models for biological sequences, as well as diagnostics and visualization functions to inspect the results. seqme can be used to evaluate both one-shot generation and iterative optimization. We show the utility of seqme by performing an antimicrobial peptide benchmark and acquiring mRNA data.</p><p><strong>Availability and implementation: </strong>seqme is released at https://github.com/szczurek-lab/seqme under the BSD 3-Clause license.</p>","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"6 1","pages":"vbag212"},"PeriodicalIF":2.6,"publicationDate":"2026-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13499220/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148802269","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":"SeizureBiomeDB: a unified database of gut microbiome alterations in pediatric and adult epilepsy.","authors":"Zehaan Walji, Parshva Dave, Meherab Ali, Omar Negmeldin, Armaan Kaur Toor, Jane Shearer, Chunlong Mu","doi":"10.1093/bioadv/vbag198","DOIUrl":"10.1093/bioadv/vbag198","url":null,"abstract":"<p><strong>Motivation: </strong>The gut microbiome plays a critical role in health, immunity, metabolism, and brain function. Growing evidence links it to neurological health via the gut-brain axis. Seizure disorders are associated with distinct microbial changes that vary by epilepsy type and developmental stage, yet data are scattered across numerous published studies. To address this, we developed SeizureBiomeDB, a curated, open-access database that compiles research on gut microbiome alterations in epilepsy across animal models and clinical populations.</p><p><strong>Results: </strong>A comprehensive manual literature search was performed across PubMed, Scopus, Embase, and the Web of Science, incorporating studies with case-control designs, animal experiments, and therapeutic interventions. Extracted data included demographics, seizure type, study design, methodology, and key findings related to the microbiome. The database was built using React and PostgreSQL for secure and efficient data management. SeizureBiomeDB integrates findings from over 80 studies covering pediatric, adult, and animal models. It is accessible via a global, searchable web platform. Interactive visualization tools, including an interactive heatmap and bar graphs, enhance data exploration and cross-study comparisons. To aid interpretation, microbial taxonomy (over 240 taxa) and function are linked to resources such as BacDive and MiMeDB. SeizureBiomeDB offers a structured hub to explore gut microbiome shifts across seizure subtypes and taxa levels. With robust filtering, users can easily extract relevant insights, fostering hypothesis generation for therapeutic targets. This platform underscores the significance of a centralized knowledge database in facilitating current advancements in gut microbiome research and epilepsy. Database URL: https://www.seizurebiome-db.ca/.</p>","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"6 1","pages":"vbag198"},"PeriodicalIF":2.6,"publicationDate":"2026-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13501535/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148815104","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":"Machine-learning-based analysis of host-depleted <i>k</i>-mer profiles for early detection and abundance estimation of coffee pathogens.","authors":"Seunghyun Lim, Ezekiel Ahn, Dapeng Zhang, Lyndel W Meinhardt, Sunchung Park","doi":"10.1093/bioadv/vbag224","DOIUrl":"10.1093/bioadv/vbag224","url":null,"abstract":"<p><strong>Motivation: </strong>Early detection of plant pathogens is essential for timely disease management, but diagnosis at low infection levels remains difficult because pathogen-derived sequences are often masked by abundant host DNA. <i>K</i>-mer-based machine learning offers a potentially sensitive, alignment-free feature-based approach for detecting infection and estimating pathogen abundance from sequencing data, but its performance across infection levels, biological backgrounds, and preprocessing strategies remains insufficiently characterized.</p><p><strong>Results: </strong>We evaluated a <i>k</i>-mer-based machine-learning framework using simulated Illumina short-read data from two <i>Coffea arabica</i> cultivars (ET39 and Typica) infected with <i>Hemileia vastatrix</i> or <i>Fusarium xylarioides</i> across a broad range of infection rates. Among ten models tested on raw and host-depleted <i>k</i>-mer profiles, logistic regression (LR) and linear support vector machine (LSVM) were the most sensitive. Host depletion with KrakenUniq substantially improved low-infection-rate detection, lowering the practical detection threshold to 0.05%. Models trained at lower infection rates performed better when tested across different infection rates than models trained at higher rates, although performance declined when the test infection rate fell below 0.05% because infected samples were increasingly misclassified as healthy. External validation showed limited biological transferability: low-infection-rate performance depended strongly on host background, whereas high-infection-rate performance depended more on pathogen identity. To improve weak-signal detection, we developed a two-stage framework combining elastic-net logistic regression for detection with ridge regression for infection rate prediction. This framework maintained near-perfect detection at 0.05% and above, improved detection at 0.01% under mixed-infection-rate training, and showed strong agreement between true and predicted infection rates (Spearman's <i>ρ</i> = 0.95). Predictive <i>k</i>-mers were extensively shared between LR and LSVM and showed distinct compositional differences between healthy- and infected-associated features.</p><p><strong>Availability and implementation: </strong>Code is available at GitHub (https://github.com/TropicalBreeding/coffee-pathogen-kmer-ml).</p>","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"6 1","pages":"vbag224"},"PeriodicalIF":2.6,"publicationDate":"2026-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13499227/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148802289","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}