{"title":"DeepPredict:一个最先进的蛋白质二级结构和相对溶剂可及性预测的web服务器。","authors":"Wafa Alanazi, Di Meng, Gianluca Pollastri","doi":"10.3389/fbinf.2025.1607402","DOIUrl":null,"url":null,"abstract":"<p><p>DeepPredict is a freely accessible web server that integrates Porter6 and PaleAle6, two state-of-the-art deep learning models designed for protein secondary structure prediction (PSSP) and relative solvent accessibility (RSA) prediction, respectively. Built on an advanced deep learning framework, DeepPredict leverages pre-trained protein language models (PLMs), specifically ESM-2, to eliminate the need for multiple sequence alignments (MSAs), enabling rapid and accurate predictions. Compared to existing methods, DeepPredict outperforms in both PSSP and RSA prediction tasks, delivering state-of-the-art performance. The server offers a user-friendly interface, catering to both computational biologists and experimental researchers. DeepPredict is available at [ https://pcrgwd.ucd.ie/wafa/] with comprehensive online documentation and downloadable example datasets.</p>","PeriodicalId":73066,"journal":{"name":"Frontiers in bioinformatics","volume":"5 ","pages":"1607402"},"PeriodicalIF":3.9000,"publicationDate":"2025-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12179536/pdf/","citationCount":"0","resultStr":"{\"title\":\"DeepPredict: a state-of-the-art web server for protein secondary structure and relative solvent accessibility prediction.\",\"authors\":\"Wafa Alanazi, Di Meng, Gianluca Pollastri\",\"doi\":\"10.3389/fbinf.2025.1607402\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>DeepPredict is a freely accessible web server that integrates Porter6 and PaleAle6, two state-of-the-art deep learning models designed for protein secondary structure prediction (PSSP) and relative solvent accessibility (RSA) prediction, respectively. Built on an advanced deep learning framework, DeepPredict leverages pre-trained protein language models (PLMs), specifically ESM-2, to eliminate the need for multiple sequence alignments (MSAs), enabling rapid and accurate predictions. Compared to existing methods, DeepPredict outperforms in both PSSP and RSA prediction tasks, delivering state-of-the-art performance. The server offers a user-friendly interface, catering to both computational biologists and experimental researchers. DeepPredict is available at [ https://pcrgwd.ucd.ie/wafa/] with comprehensive online documentation and downloadable example datasets.</p>\",\"PeriodicalId\":73066,\"journal\":{\"name\":\"Frontiers in bioinformatics\",\"volume\":\"5 \",\"pages\":\"1607402\"},\"PeriodicalIF\":3.9000,\"publicationDate\":\"2025-06-06\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12179536/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Frontiers in bioinformatics\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.3389/fbinf.2025.1607402\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"2025/1/1 0:00:00\",\"PubModel\":\"eCollection\",\"JCR\":\"Q2\",\"JCRName\":\"MATHEMATICAL & COMPUTATIONAL BIOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Frontiers in bioinformatics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.3389/fbinf.2025.1607402","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/1/1 0:00:00","PubModel":"eCollection","JCR":"Q2","JCRName":"MATHEMATICAL & COMPUTATIONAL BIOLOGY","Score":null,"Total":0}
DeepPredict: a state-of-the-art web server for protein secondary structure and relative solvent accessibility prediction.
DeepPredict is a freely accessible web server that integrates Porter6 and PaleAle6, two state-of-the-art deep learning models designed for protein secondary structure prediction (PSSP) and relative solvent accessibility (RSA) prediction, respectively. Built on an advanced deep learning framework, DeepPredict leverages pre-trained protein language models (PLMs), specifically ESM-2, to eliminate the need for multiple sequence alignments (MSAs), enabling rapid and accurate predictions. Compared to existing methods, DeepPredict outperforms in both PSSP and RSA prediction tasks, delivering state-of-the-art performance. The server offers a user-friendly interface, catering to both computational biologists and experimental researchers. DeepPredict is available at [ https://pcrgwd.ucd.ie/wafa/] with comprehensive online documentation and downloadable example datasets.