{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,31]],"date-time":"2025-12-31T10:56:07Z","timestamp":1767178567960,"version":"build-2238731810"},"update-to":[{"DOI":"10.1371\/journal.pcbi.1013622","type":"new_version","label":"New version","source":"publisher","updated":{"date-parts":[[2025,10,31]],"date-time":"2025-10-31T00:00:00Z","timestamp":1761868800000}}],"reference-count":35,"publisher":"Public Library of Science (PLoS)","issue":"10","license":[{"start":{"date-parts":[[2025,10,27]],"date-time":"2025-10-27T00:00:00Z","timestamp":1761523200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Science and Technology Program of Guangzhou","award":["2024A03J0464"],"award-info":[{"award-number":["2024A03J0464"]}]},{"name":"Guangdong University Featured Innovation Program Project","award":["Grant No. 2023KTSCX256"],"award-info":[{"award-number":["Grant No. 2023KTSCX256"]}]},{"name":"National Key Laboratory of Data Space Technology and System","award":["Grant No.QZQC2024004-6"],"award-info":[{"award-number":["Grant No.QZQC2024004-6"]}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62572215"],"award-info":[{"award-number":["62572215"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.ploscompbiol.org"],"crossmark-restriction":false},"short-container-title":["PLoS Comput Biol"],"abstract":"<jats:p>Accurately predicting the functions of multi-functional therapeutic peptides is crucial for the development of related drugs. However, existing peptide function prediction methods largely rely on either a single type of feature or a single model architecture, limiting prediction accuracy and applicability. Additionally, training better-performing models on datasets with class imbalance issues remains a significant challenge. In this study, we propose the multi-functional therapeutic peptide of multi-feature fusion prediction (MFTP_MFFP) model, a novel method for predicting the functionality of multi-functional therapeutic peptides. This approach uses various encoding techniques to process peptide sequence data, generating multiple features that help the model learn hidden information within the sequences. To maximize the effectiveness of these features, we propose a gated feature fusion module that efficiently integrates them. The module assigns learnable gating weights to each feature, optimizing integration and enhancing fusion efficiency. The fused features are then passed into a neural network model for feature extraction. Additionally, we propose a marginal focal dice loss function (MFDL) to address the class imbalance and improve the model\u2019s prediction performance. Experimental results show that the MFTP_MFFP model outperforms existing models in all evaluation metrics, demonstrating its robustness and effectiveness in multi-functional therapeutic peptide prediction tasks.<\/jats:p>","DOI":"10.1371\/journal.pcbi.1013622","type":"journal-article","created":{"date-parts":[[2025,10,27]],"date-time":"2025-10-27T17:51:40Z","timestamp":1761587500000},"page":"e1013622","update-policy":"https:\/\/doi.org\/10.1371\/journal.pcbi.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Multi-feature fusion network with marginal focal dice loss for multi-label therapeutic peptide prediction"],"prefix":"10.1371","volume":"21","author":[{"given":"Yijun","family":"Mao","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-2537-2639","authenticated-orcid":true,"given":"Yurong","family":"Weng","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jian","family":"Weng","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ming","family":"Li","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Wanrong","family":"Gu","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Rui","family":"Pang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xudong","family":"Lin","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yunyan","family":"Xiong","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1782-6007","authenticated-orcid":true,"given":"Deyu","family":"Tang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"340","published-online":{"date-parts":[[2025,10,27]]},"reference":[{"issue":"13","key":"pcbi.1013622.ref001","doi-asserted-by":"crossref","first-page":"2273","DOI":"10.1080\/10408398.2017.1319795","article-title":"Current trends and perspectives of bioactive peptides","volume":"58","author":"EB-M Daliri","year":"2018","journal-title":"Crit Rev Food Sci Nutr."},{"issue":"1","key":"pcbi.1013622.ref002","doi-asserted-by":"crossref","first-page":"48","DOI":"10.1038\/s41392-022-00904-4","article-title":"Therapeutic peptides: current applications and future directions","volume":"7","author":"L Wang","year":"2022","journal-title":"Signal Transduct Target Ther."},{"key":"pcbi.1013622.ref003","article-title":"APD3: the antimicrobial peptide database as a tool for research and education","volume":"44","author":"G Wang","year":"2016","journal-title":"Nucleic Acids Res."},{"issue":"2","key":"pcbi.1013622.ref004","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0190134","article-title":"TopicalPdb: a database of topically delivered peptides","volume":"13","author":"D Mathur","year":"2018","journal-title":"PLoS One."},{"issue":"3","key":"pcbi.1013622.ref005","doi-asserted-by":"crossref","first-page":"426","DOI":"10.1007\/s12539-021-00436-5","article-title":"TUPDB: target-unrelated peptide data bank","volume":"13","author":"B He","year":"2021","journal-title":"Interdiscip Sci."},{"issue":"4","key":"pcbi.1013622.ref006","article-title":"iAMPCN: a deep-learning approach for identifying antimicrobial peptides and their functional activities","volume":"24","author":"J Xu","year":"2023","journal-title":"Brief Bioinform."},{"issue":"9","key":"pcbi.1013622.ref007","article-title":"PrMFTP: multi-functional therapeutic peptides prediction based on multi-head self-attention mechanism and class weight optimization","volume":"18","author":"W Yan","year":"2022","journal-title":"PLoS Comput Biol."},{"key":"pcbi.1013622.ref008","article-title":"Focal loss for dense object detection","author":"T Lin","year":"2017","journal-title":"arXiv preprint"},{"key":"pcbi.1013622.ref009","doi-asserted-by":"crossref","unstructured":"Li X, Sun X, Meng Y, Liang J, Wu F, Li J. Dice loss for data-imbalanced NLP tasks. arXiv preprint 2019. https:\/\/arxiv.org\/abs\/1911.02855","DOI":"10.18653\/v1\/2020.acl-main.45"},{"issue":"6","key":"pcbi.1013622.ref010","article-title":"Deep learning-based multi-functional therapeutic peptides prediction with a multi-label focal dice loss function","volume":"39","author":"H Fan","year":"2023","journal-title":"Bioinformatics."},{"issue":"20","key":"pcbi.1013622.ref011","first-page":"10","article-title":"Graph attention networks","volume":"1050","author":"P Velickovic","year":"2017","journal-title":"Stat."},{"key":"pcbi.1013622.ref012","article-title":"Learning imbalanced datasets with label-distribution-aware margin loss","volume":"32","author":"K Cao","year":"2019","journal-title":"Advances in Neural Information Processing Systems."},{"issue":"2","key":"pcbi.1013622.ref013","doi-asserted-by":"crossref","first-page":"137","DOI":"10.1007\/s11222-009-9153-8","article-title":"Estimation of prediction error by using K-fold cross-validation","volume":"21","author":"T Fushiki","year":"2009","journal-title":"Stat Comput."},{"key":"pcbi.1013622.ref014","unstructured":"Staudemeyer RC, Morris ER. Understanding LSTM\u2013a tutorial into long short-term memory recurrent neural networks. arXiv preprint 2019. https:\/\/arxiv.org\/abs\/1909.09586"},{"key":"pcbi.1013622.ref015","doi-asserted-by":"crossref","unstructured":"Siami-Namini S, Tavakoli N, Namin AS. The performance of LSTM and BiLSTM in forecasting time series. In: 2019 IEEE International conference on big data (Big Data). 2019. p. 3285\u201392.","DOI":"10.1109\/BigData47090.2019.9005997"},{"issue":"1","key":"pcbi.1013622.ref016","doi-asserted-by":"crossref","first-page":"374","DOI":"10.1093\/nar\/28.1.374","article-title":"AAindex: amino acid index database","volume":"28","author":"S Kawashima","year":"2000","journal-title":"Nucleic Acids Res."},{"key":"pcbi.1013622.ref017","doi-asserted-by":"crossref","DOI":"10.1093\/nar\/gkac351","article-title":"iFeatureOmega: an integrative platform for engineering, visualization and analysis of features from molecular sequences, structural and ligand data sets","volume":"50","author":"Z Chen","year":"2022","journal-title":"Nucleic Acids Res."},{"key":"pcbi.1013622.ref018","doi-asserted-by":"crossref","unstructured":"Ridnik T, Ben-Baruch E, Zamir N, Noy A, Friedman I, Protter M, et al. Asymmetric loss for multi-label classification. In: 2021 IEEE\/CVF International Conference on Computer Vision (ICCV). 2021. p. 82\u201391. https:\/\/doi.org\/10.1109\/iccv48922.2021.00015","DOI":"10.1109\/ICCV48922.2021.00015"},{"issue":"6","key":"pcbi.1013622.ref019","doi-asserted-by":"crossref","first-page":"707","DOI":"10.3390\/ph15060707","article-title":"MPMABP: a CNN and Bi-LSTM-based method for predicting multi-activities of bioactive peptides","volume":"15","author":"Y Li","year":"2022","journal-title":"Pharmaceuticals (Basel)."},{"issue":"1","key":"pcbi.1013622.ref020","doi-asserted-by":"crossref","DOI":"10.1093\/bib\/bbab414","article-title":"Identifying multi-functional bioactive peptide functions using multi-label deep learning","volume":"23","author":"W Tang","year":"2022","journal-title":"Brief Bioinform."},{"issue":"12","key":"pcbi.1013622.ref021","doi-asserted-by":"crossref","first-page":"2961","DOI":"10.1021\/acs.jcim.2c00526","article-title":"Sequential properties representation scheme for recurrent neural network-based prediction of therapeutic peptides","volume":"62","author":"E Otovi\u0107","year":"2022","journal-title":"J Chem Inf Model."},{"issue":"1","key":"pcbi.1013622.ref022","doi-asserted-by":"crossref","first-page":"1","DOI":"10.2174\/0115748936299646240625092734","article-title":"MFTP-Tool: a wide & deep learning framework for multi-functional therapeutic peptides prediction","volume":"20","author":"Y Lv","year":"2025","journal-title":"CBIO."},{"key":"pcbi.1013622.ref023","unstructured":"Vaswani A. Attention is all you need. Advances in Neural Information Processing Systems. 2017."},{"issue":"3","key":"pcbi.1013622.ref024","doi-asserted-by":"crossref","first-page":"300","DOI":"10.1038\/s42256-022-00459-7","article-title":"A transformer-based model to predict peptide\u2013HLA class I binding and optimize mutated peptides for vaccine design","volume":"4","author":"Y Chu","year":"2022","journal-title":"Nat Mach Intell."},{"issue":"1","key":"pcbi.1013622.ref025","doi-asserted-by":"crossref","first-page":"94","DOI":"10.1109\/TFUZZ.2024.3369944","article-title":"A novel centralized federated deep fuzzy neural network with multi-objectives neural architecture search for epistatic detection","volume":"33","author":"X Wu","year":"2025","journal-title":"IEEE Trans Fuzzy Syst."},{"issue":"6","key":"pcbi.1013622.ref026","doi-asserted-by":"crossref","first-page":"7135","DOI":"10.1016\/j.eswa.2010.12.048","article-title":"Using Gaussian membership functions for improving the reliability and robustness of students\u2019 evaluation systems","volume":"38","author":"IA Hameed","year":"2011","journal-title":"Expert Systems with Applications."},{"issue":"10","key":"pcbi.1013622.ref027","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0046618","article-title":"L-ornithine derived polyamines in cystic fibrosis airways","volume":"7","author":"H Grasemann","year":"2012","journal-title":"PLoS One."},{"issue":"3","key":"pcbi.1013622.ref028","article-title":"DeepSS2GO: protein function prediction from secondary structure","volume":"25","author":"FV Song","year":"2024","journal-title":"Brief Bioinform."},{"issue":"1","key":"pcbi.1013622.ref029","doi-asserted-by":"crossref","first-page":"137","DOI":"10.3390\/biom13010137","article-title":"Prediction of protein function from tertiary structure of the active site in heme proteins by convolutional neural network","volume":"13","author":"HX Kondo","year":"2023","journal-title":"Biomolecules."},{"issue":"1","key":"pcbi.1013622.ref030","doi-asserted-by":"crossref","first-page":"4175","DOI":"10.1038\/s41467-023-39909-0","article-title":"Discovering functionally important sites in proteins","volume":"14","author":"M Cagiada","year":"2023","journal-title":"Nat Commun."},{"key":"pcbi.1013622.ref031","doi-asserted-by":"crossref","DOI":"10.7717\/peerj.6830","article-title":"FunPred 3.0: improved protein function prediction using protein interaction network","volume":"7","author":"S Saha","year":"2019","journal-title":"PeerJ."},{"issue":"1","key":"pcbi.1013622.ref032","doi-asserted-by":"crossref","first-page":"4","DOI":"10.1109\/TNNLS.2020.2978386","article-title":"A comprehensive survey on graph neural networks","volume":"32","author":"Z Wu","year":"2021","journal-title":"IEEE Trans Neural Netw Learn Syst."},{"key":"pcbi.1013622.ref033","doi-asserted-by":"crossref","unstructured":"Han J, Moraga C. The influence of the sigmoid function parameters on the speed of backpropagation learning. In: International workshop on artificial neural networks. 1995. p. 195\u2013201.","DOI":"10.1007\/3-540-59497-3_175"},{"key":"pcbi.1013622.ref034","doi-asserted-by":"crossref","unstructured":"Mirjalili S, Mirjalili S. Genetic algorithm. Evolutionary algorithms and neural networks: theory and applications. 2019. p. 43\u201355.","DOI":"10.1007\/978-3-319-93025-1_4"},{"key":"pcbi.1013622.ref035","unstructured":"Xu B. Empirical evaluation of rectified activations in convolutional network. arXiv preprint 2015. https:\/\/arxiv.org\/abs\/1505.00853"}],"updated-by":[{"DOI":"10.1371\/journal.pcbi.1013622","type":"new_version","label":"New version","source":"publisher","updated":{"date-parts":[[2025,10,31]],"date-time":"2025-10-31T00:00:00Z","timestamp":1761868800000}}],"container-title":["PLOS Computational Biology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pcbi.1013622","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,31]],"date-time":"2025-10-31T18:08:52Z","timestamp":1761934132000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pcbi.1013622"}},"subtitle":[],"editor":[{"given":"Jinshan","family":"Xu","sequence":"first","affiliation":[],"role":[{"role":"editor","vocabulary":"crossref"}]}],"short-title":[],"issued":{"date-parts":[[2025,10,27]]},"references-count":35,"journal-issue":{"issue":"10","published-online":{"date-parts":[[2025,10,27]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pcbi.1013622","relation":{},"ISSN":["1553-7358"],"issn-type":[{"value":"1553-7358","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,10,27]]}}}