{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,7]],"date-time":"2026-04-07T14:50:16Z","timestamp":1775573416547,"version":"3.50.1"},"reference-count":63,"publisher":"Oxford University Press (OUP)","issue":"5","license":[{"start":{"date-parts":[[2025,9,22]],"date-time":"2025-09-22T00:00:00Z","timestamp":1758499200000},"content-version":"vor","delay-in-days":22,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["32202891"],"award-info":[{"award-number":["32202891"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,8,31]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Accurate prediction of bacterial virulence factors (VFs) is crucial for combating infectious diseases, yet traditional methods often fail to capture their complex sequence properties. We address this challenge by leveraging deep, context-aware representations from large-scale protein language models (PLMs). Our framework begins with a systematic engineering of features from ESM-2 and ProtT5, which confirmed their complementary nature but also revealed that simple concatenation is a suboptimal fusion strategy due to a \u201cfeature overshadowing\u201d effect. To overcome this, we developed two novel architectures: VF-Iter, for robust feature enhancement via iterative low-rank updates, and the Dual-Path Feature Fusion (DPF) network, for intelligently integrating the complementary embeddings. The construction of our final model, VF-Fuse, involved a two-stage process. First, we selected four powerful and diverse base models representing our distinct feature strategies (ESM-2 only, ProtT5 only, simple concatenation, and DPF). Second, we empirically determined the best method for combining their predictions by benchmarking 15 ensemble techniques, from which Majority Voting emerged as the superior choice. On the independent test set, VF-Fuse establishes a new state of the art, achieving a superior F1-Score of 87.15% and a Matthews Correlation Coefficient of 73.61%. This F1-Score marks a significant 3.3% improvement over the previous best method, driven by an excellent balance between a high Sensitivity of 90.1% and a strong Specificity of 83.33%. Crucially, in-depth interpretability analyses validated our architectural design, demonstrating how the DPF model learns to intelligently route complementary features to specialized pathways.<\/jats:p>","DOI":"10.1093\/bib\/bbaf481","type":"journal-article","created":{"date-parts":[[2025,8,29]],"date-time":"2025-08-29T11:41:14Z","timestamp":1756467674000},"source":"Crossref","is-referenced-by-count":1,"title":["VF-Fuse: a dual-path feature fusion and iterative update architecture for virulence factor prediction"],"prefix":"10.1093","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0009-0002-8395-8574","authenticated-orcid":false,"given":"Liang","family":"Huang","sequence":"first","affiliation":[{"name":"School of Information and Artificial Intelligence, Anhui Agricultural University , Changjiang West Road 130, Shushan District, Hefei 230036, Anhui Province ,","place":["China"]},{"name":"Anhui Province Key Laboratory of Smart Agricultural Technology and Equipment , Changjiang West Road 130, Shushan District, Hefei 230036, Anhui Province ,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiangyu","family":"Yu","sequence":"additional","affiliation":[{"name":"School of Information and Artificial Intelligence, Anhui Agricultural University , Changjiang West Road 130, Shushan District, Hefei 230036, Anhui Province ,","place":["China"]},{"name":"Anhui Province Key Laboratory of Smart Agricultural Technology and Equipment , Changjiang West Road 130, Shushan District, Hefei 230036, Anhui Province ,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shumei","family":"Li","sequence":"additional","affiliation":[{"name":"School of Information and Artificial Intelligence, Anhui Agricultural University , Changjiang West Road 130, Shushan District, Hefei 230036, Anhui Province ,","place":["China"]},{"name":"Anhui Province Key Laboratory of Smart Agricultural Technology and Equipment , Changjiang West Road 130, Shushan District, Hefei 230036, Anhui Province ,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Qingwei","family":"Chen","sequence":"additional","affiliation":[{"name":"School of Information and Artificial Intelligence, Anhui Agricultural University , Changjiang West Road 130, Shushan District, Hefei 230036, Anhui Province ,","place":["China"]},{"name":"Anhui Province Key Laboratory of Smart Agricultural Technology and Equipment , Changjiang West Road 130, Shushan District, Hefei 230036, Anhui Province ,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Dan","family":"Xu","sequence":"additional","affiliation":[{"name":"School of Information and Artificial Intelligence, Anhui Agricultural University , Changjiang West Road 130, Shushan District, Hefei 230036, Anhui Province ,","place":["China"]},{"name":"Anhui Province Key Laboratory of Smart Agricultural Technology and Equipment , Changjiang West Road 130, Shushan District, Hefei 230036, Anhui Province ,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhao","family":"Qi","sequence":"additional","affiliation":[{"name":"School of Information and Artificial Intelligence, Anhui Agricultural University , Changjiang West Road 130, Shushan District, Hefei 230036, Anhui Province ,","place":["China"]},{"name":"Anhui Province Key Laboratory of Smart Agricultural Technology and Equipment , Changjiang West Road 130, Shushan District, Hefei 230036, Anhui Province ,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2025,9,22]]},"reference":[{"key":"2025092202361151500_ref1","doi-asserted-by":"publisher","first-page":"892","DOI":"10.1093\/femsre\/fuv029","article-title":"Targeted imaging of bacterial infections: Advances, hurdles and hopes","volume":"39","author":"Oosten","year":"2015","journal-title":"FEMS Microbiol Rev"},{"key":"2025092202361151500_ref2","doi-asserted-by":"publisher","first-page":"17","DOI":"10.1038\/nrmicro732","article-title":"The damage-response framework of microbial pathogenesis","volume":"1","author":"Casadevall","year":"2003","journal-title":"Nat Rev Microbiol"},{"key":"2025092202361151500_ref3","doi-asserted-by":"publisher","first-page":"363","DOI":"10.1038\/nrmicro1636","article-title":"The tad locus: Postcards from the widespread colonization island","volume":"5","author":"Tomich","year":"2007","journal-title":"Nat Rev Microbiol"},{"key":"2025092202361151500_ref4","doi-asserted-by":"publisher","first-page":"93","DOI":"10.1016\/j.cbpa.2008.01.023","article-title":"Discovery of virulence factors of pathogenic bacteria","volume":"12","author":"Wu","year":"2008","journal-title":"Curr Opin Chem Biol"},{"key":"2025092202361151500_ref5","doi-asserted-by":"publisher","first-page":"D912","DOI":"10.1093\/nar\/gkab1107","article-title":"VFDB 2022: A general classification scheme for bacterial virulence factors","volume":"50","author":"Liu","year":"2022","journal-title":"Nucleic Acids Res"},{"key":"2025092202361151500_ref6","doi-asserted-by":"publisher","first-page":"D693","DOI":"10.1093\/nar\/gky999","article-title":"Victors: A web-based knowledge base of virulence factors in human and animal pathogens","volume":"47","author":"Sayers","year":"2019","journal-title":"Nucleic Acids Res"},{"key":"2025092202361151500_ref7","doi-asserted-by":"publisher","first-page":"D391","DOI":"10.1093\/nar\/gkl791","article-title":"MvirDB\u2013a microbial database of protein toxins, virulence factors and antibiotic resistance genes for bio-defence applications","volume":"35","author":"Zhou","year":"2007","journal-title":"Nucleic Acids Res"},{"key":"2025092202361151500_ref8","doi-asserted-by":"publisher","first-page":"W6","DOI":"10.1093\/nar\/gkl164","article-title":"BLAST: Improvements for better sequence analysis","volume":"34","author":"Ye","year":"2006","journal-title":"Nucleic Acids Res"},{"key":"2025092202361151500_ref9","doi-asserted-by":"publisher","first-page":"D687","DOI":"10.1093\/nar\/gky1080","article-title":"VFDB 2019: A comparative pathogenomic platform with an interactive web interface","volume":"47","author":"Liu","year":"2019","journal-title":"Nucleic Acids Res"},{"key":"2025092202361151500_ref10","doi-asserted-by":"publisher","first-page":"W768","DOI":"10.1093\/nar\/gkac321","article-title":"VRprofile2: Detection of antibiotic resistance-associated mobilome in bacterial pathogens","volume":"50","author":"Wang","year":"2022","journal-title":"Nucleic Acids Res"},{"key":"2025092202361151500_ref11","doi-asserted-by":"publisher","first-page":"566","DOI":"10.1093\/bib\/bbw141","article-title":"VRprofile: Gene-cluster-detection-based profiling of virulence and antibiotic resistance traits encoded within genome sequences of pathogenic bacteria","volume":"19","author":"Li","year":"2017","journal-title":"Brief Bioinform"},{"key":"2025092202361151500_ref12","doi-asserted-by":"publisher","DOI":"10.1186\/1471-2105-9-62","article-title":"VirulentPred: A SVM based prediction method for virulent proteins in bacterial pathogens","volume":"9","author":"Garg","year":"2008","journal-title":"BMC Bioinformatics"},{"key":"2025092202361151500_ref13","doi-asserted-by":"publisher","first-page":"e93907","DOI":"10.1371\/journal.pone.0093907","article-title":"MP3: A software tool for the prediction of pathogenic proteins in genomic and metagenomic data","volume":"9","author":"Gupta","year":"2014","journal-title":"PLoS One"},{"key":"2025092202361151500_ref14","doi-asserted-by":"publisher","first-page":"1596","DOI":"10.1093\/bib\/bbz076","article-title":"Predicting bacterial virulence factors - evaluation of machine learning and negative data strategies","volume":"21","author":"Rentzsch","year":"2020","journal-title":"Brief Bioinform"},{"key":"2025092202361151500_ref15","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbaa125","article-title":"DeepVF: A deep learning-based hybrid framework for identifying virulence factors using the stacking strategy","volume":"22","author":"Xie","year":"2020","journal-title":"Brief Bioinform"},{"key":"2025092202361151500_ref16","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2023.107662","article-title":"VF-pred: Predicting virulence factor using sequence alignment percentage and ensemble learning models","volume":"168","author":"Singh","year":"2023","journal-title":"Comput Biol Med"},{"key":"2025092202361151500_ref17","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1706.03762","article-title":"Attention is all you need","volume-title":"Advances in Neural Information Processing Systems","author":"Vaswani"},{"key":"2025092202361151500_ref18","doi-asserted-by":"publisher","first-page":"1123","DOI":"10.1126\/science.ade2574","article-title":"Evolutionary-scale prediction of atomic-level protein structure with a language model","volume":"379","author":"Lin","year":"2023","journal-title":"Science"},{"key":"2025092202361151500_ref19","doi-asserted-by":"publisher","first-page":"7112","DOI":"10.1109\/tpami.2021.3095381","article-title":"ProtTrans: Toward understanding the language of life through self-supervised learning","volume":"44","author":"Elnaggar","year":"2021","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"2025092202361151500_ref20","doi-asserted-by":"publisher","first-page":"2102","DOI":"10.1093\/bioinformatics\/btac020","article-title":"ProteinBERT: A universal deep-learning model of protein sequence and function","volume":"38","author":"Brandes","year":"2022","journal-title":"Bioinformatics"},{"key":"2025092202361151500_ref21","doi-asserted-by":"publisher","first-page":"1803","DOI":"10.1002\/jcb.30490","article-title":"PortPred: Exploiting deep learning embeddings of amino acid sequences for the identification of transporter proteins and their substrates","volume":"124","author":"Anteghini","year":"2023","journal-title":"J Cell Biochem"},{"key":"2025092202361151500_ref22","volume-title":"Antifreeze Protein Recognition and Classification Model Based on BERT and Structural Domain Annotation. iScience","author":"Chen","year":"2025"},{"key":"2025092202361151500_ref23","author":"Chae","year":"2024"},{"key":"2025092202361151500_ref24","doi-asserted-by":"publisher","first-page":"108429","DOI":"10.1016\/j.compbiolchem.2025.108429","article-title":"ProAttUnet: Advancing protein secondary structure prediction with deep learning via U-net dual-pathway feature fusion and ESM2 pretrained protein language model","volume":"118","author":"Cheng","year":"2025","journal-title":"Comput Biol Chem"},{"key":"2025092202361151500_ref25","doi-asserted-by":"publisher","first-page":"125701","DOI":"10.1088\/1361-6501\/ad761c","article-title":"ESMDNN-PPI: A new protein\u2013protein interaction prediction model developed with protein language model of ESM2 and deep neural network","volume":"35","author":"Li","year":"2024","journal-title":"Meas Sci Technol"},{"key":"2025092202361151500_ref26","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbad534","article-title":"Interpretable feature extraction and dimensionality reduction in ESM2 for protein localization prediction","volume":"25","author":"Luo","year":"2024","journal-title":"Brief Bioinform"},{"key":"2025092202361151500_ref27","doi-asserted-by":"publisher","first-page":"107801","DOI":"10.1016\/j.future.2025.107801","article-title":"DrugPred: An ensemble learning model based on ESM2 for predicting potential druggable proteins","volume":"170","author":"Zhang","year":"2025","journal-title":"Futur Gener Comput Syst"},{"key":"2025092202361151500_ref28","doi-asserted-by":"publisher","first-page":"108062","DOI":"10.1016\/j.compbiomed.2024.108062","article-title":"Identifying virulence factors using graph transformer autoencoder with ESMFold-predicted structures","volume":"170","author":"Li","year":"2024","journal-title":"Comput Biol Med"},{"key":"2025092202361151500_ref29","doi-asserted-by":"publisher","DOI":"10.3390\/genes15091170","article-title":"DTVF: A user-friendly tool for virulence factor prediction based on ProtT5 and deep transfer learning models","volume":"15","author":"Sun","year":"2024","journal-title":"Genes"},{"key":"2025092202361151500_ref30","author":"Sun","year":"2024"},{"key":"2025092202361151500_ref31"},{"key":"2025092202361151500_ref32","author":"Yang","year":"2023"},{"key":"2025092202361151500_ref33","author":"Yang","year":"2024"},{"key":"2025092202361151500_ref34","author":"Yang","year":"2024"},{"key":"2025092202361151500_ref35","author":"Siems","year":"2025"},{"key":"2025092202361151500_ref36","doi-asserted-by":"publisher","first-page":"D401","DOI":"10.1093\/nar\/gkl858","article-title":"PATRIC: The VBI PathoSystems resource integration Center","volume":"35","author":"Snyder","year":"2007","journal-title":"Nucleic Acids Res"},{"key":"2025092202361151500_ref37","doi-asserted-by":"publisher","first-page":"680","DOI":"10.1093\/bioinformatics\/btq003","article-title":"CD-HIT suite: A web server for clustering and comparing biological sequences","volume":"26","author":"Huang","year":"2010","journal-title":"Bioinformatics"},{"key":"2025092202361151500_ref38","doi-asserted-by":"publisher","first-page":"270","DOI":"10.1093\/bib\/bbw021","article-title":"Prediction and identification of the effectors of heterotrimeric G proteins in rice (Oryza sativa L.)","volume":"18","author":"Li","year":"2016","journal-title":"Brief Bioinform"},{"key":"2025092202361151500_ref39","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbx164","article-title":"Systematic analysis and prediction of type IV secreted effector proteins by machine learning approaches","volume":"20","author":"Wang","year":"2019","journal-title":"Brief Bioinform"},{"key":"2025092202361151500_ref40","doi-asserted-by":"publisher","first-page":"648","DOI":"10.1089\/omi.2015.0095","article-title":"Harnessing computational biology for exact linear B-cell epitope prediction: A novel amino acid composition-based feature descriptor. OMICS: A journal of","volume":"19","author":"Saravanan","year":"2015","journal-title":"Integr Biol"},{"key":"2025092202361151500_ref41","doi-asserted-by":"publisher","first-page":"386","DOI":"10.1016\/j.ab.2007.10.012","article-title":"PseAAC: A flexible web server for generating various kinds of protein pseudo amino acid composition","volume":"373","author":"Shen","year":"2008","journal-title":"Anal Biochem"},{"key":"2025092202361151500_ref42","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1002\/prot.1072","article-title":"Prediction of protein cellular attributes using pseudo-amino acid composition","volume":"44","author":"Chou","year":"2001","journal-title":"Proteins Struct Funct Bioinf"},{"key":"2025092202361151500_ref43","doi-asserted-by":"publisher","first-page":"477","DOI":"10.1006\/bbrc.2000.3815","article-title":"Prediction of protein subcellular locations by incorporating quasi-sequence-order effect","volume":"278","author":"Chou","year":"2000","journal-title":"Biochem Biophys Res Commun"},{"key":"2025092202361151500_ref44","doi-asserted-by":"publisher","DOI":"10.3390\/molecules29204965","article-title":"Protein A-like peptide design based on diffusion and ESM2 models","volume":"29","author":"Zhao","year":"2024","journal-title":"Molecules"},{"key":"2025092202361151500_ref45","doi-asserted-by":"publisher","first-page":"1315","DOI":"10.1038\/s41592-019-0598-1","article-title":"Unified rational protein engineering with sequence-based deep representation learning","volume":"16","author":"Alley","year":"2019","journal-title":"Nat Methods"},{"key":"2025092202361151500_ref46","author":"Rao","year":"2019"},{"key":"2025092202361151500_ref47","article-title":"Rectifier nonlinearities improve neural network acoustic models","volume":"28","author":"Maas","journal-title":"Proceedings of the 30th International Conference on Machine Learning"},{"key":"2025092202361151500_ref48","first-page":"1929","article-title":"Dropout: A simple way to prevent neural networks from overfitting","volume":"15","author":"Srivastava","year":"2014","journal-title":"J Mach Learn Res"},{"key":"2025092202361151500_ref49","author":"Hendrycks","year":"2016"},{"key":"2025092202361151500_ref50","author":"Ioffe","year":"2015"},{"key":"2025092202361151500_ref51","doi-asserted-by":"crossref","volume-title":"Memristor-Based Circuit Demonstration of Gated Recurrent Unit for Predictable Neural Network","author":"Zhang","DOI":"10.1109\/TED.2022.3217116"},{"key":"2025092202361151500_ref52","doi-asserted-by":"publisher","DOI":"10.34133\/research.0153","article-title":"Enzyme commission number prediction and benchmarking with hierarchical dual-core multitask learning framework","volume":"6","author":"Shi","year":"2023","journal-title":"Research"},{"key":"2025092202361151500_ref53","doi-asserted-by":"publisher","first-page":"3115","DOI":"10.1007\/s10462-020-09922-6","article-title":"Ensemble learning based on random super-reduct and resampling","volume":"54","author":"Jiang","year":"2020","journal-title":"Artif Intell Rev"},{"key":"2025092202361151500_ref54","doi-asserted-by":"publisher","first-page":"5545","DOI":"10.1007\/s10462-022-10283-5","article-title":"A survey on ensemble learning under the era of deep learning","volume":"56","author":"Yang","year":"2022","journal-title":"Artif Intell Rev"},{"key":"2025092202361151500_ref55","author":"Liaw","year":"2018"},{"key":"2025092202361151500_ref56","volume-title":"KDD","author":"Akiba"},{"key":"2025092202361151500_ref57","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbaf274","article-title":"MultiPep-DLCL: Recognition of multifunctional therapeutic peptides through deep learning with label-sequence contrastive learning","volume":"26","author":"Li","year":"2025","journal-title":"Brief Bioinform"},{"key":"2025092202361151500_ref58","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btad334","article-title":"Deep learning-based multi-functional therapeutic peptides prediction with a multi-label focal dice loss function","volume":"39","author":"Fan","year":"2023","journal-title":"Bioinformatics"},{"key":"2025092202361151500_ref59","doi-asserted-by":"publisher","DOI":"10.1007\/s11432-024-4147-8","article-title":"TPpred-SC: Multi-functional therapeutic peptide prediction based on multi-label supervised contrastive learning","volume":"67","author":"Yan","year":"2024","journal-title":"SCIENCE CHINA Inf Sci"},{"key":"2025092202361151500_ref60","doi-asserted-by":"publisher","first-page":"2431","DOI":"10.1093\/bioinformatics\/btr411","article-title":"Simple high-throughput annotation pipeline (SHAP)","volume":"27","author":"DeMaere","year":"2011","journal-title":"Bioinformatics"},{"key":"2025092202361151500_ref61","author":"Linderman","year":"2017"},{"key":"2025092202361151500_ref62","doi-asserted-by":"publisher","DOI":"10.3390\/sym16040464","article-title":"UniproLcad: Accurate identification of antimicrobial peptide by fusing multiple pre-trained protein language models","volume":"16","author":"Wang","year":"2024","journal-title":"Symmetry"},{"key":"2025092202361151500_ref63","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbab083","article-title":"Comprehensive assessment of machine learning-based methods for predicting antimicrobial peptides","volume":"22","author":"Xu","year":"2021","journal-title":"Brief Bioinform"}],"container-title":["Briefings in Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/26\/5\/bbaf481\/64339210\/bbaf481.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/26\/5\/bbaf481\/64339210\/bbaf481.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,22]],"date-time":"2025-09-22T06:36:19Z","timestamp":1758522979000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bib\/article\/doi\/10.1093\/bib\/bbaf481\/8260786"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,8,31]]},"references-count":63,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2025,8,31]]}},"URL":"https:\/\/doi.org\/10.1093\/bib\/bbaf481","relation":{},"ISSN":["1467-5463","1477-4054"],"issn-type":[{"value":"1467-5463","type":"print"},{"value":"1477-4054","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2025,9]]},"published":{"date-parts":[[2025,8,31]]},"article-number":"bbaf481"}}