Hossein Ebrahimikondori, Darcy Sutherland, Anat Yanai, Amelia Richter, Ali Salehi, Chenkai Li, Lauren Coombe, Monica Kotkoff, René L. Warren, Inanc Birol
{"title":"用于多肽毒性预测的结构感知深度学习模型","authors":"Hossein Ebrahimikondori, Darcy Sutherland, Anat Yanai, Amelia Richter, Ali Salehi, Chenkai Li, Lauren Coombe, Monica Kotkoff, René L. Warren, Inanc Birol","doi":"10.1002/pro.5076","DOIUrl":null,"url":null,"abstract":"Antimicrobial resistance is a critical public health concern, necessitating the exploration of alternative treatments. While antimicrobial peptides (AMPs) show promise, assessing their toxicity using traditional wet lab methods is both time‐consuming and costly. We introduce tAMPer, a novel multi‐modal deep learning model designed to predict peptide toxicity by integrating the underlying amino acid sequence composition and the three‐dimensional structure of peptides. tAMPer adopts a graph‐based representation for peptides, encoding ColabFold‐predicted structures, where nodes represent amino acids and edges represent spatial interactions. Structural features are extracted using graph neural networks, and recurrent neural networks capture sequential dependencies. tAMPer's performance was assessed on a publicly available protein toxicity benchmark and an AMP hemolysis data we generated. On the latter, tAMPer achieves an F1‐score of 68.7%, outperforming the second‐best method by 23.4%. On the protein benchmark, tAMPer exhibited an improvement of over 3.0% in the F1‐score compared to current state‐of‐the‐art methods. We anticipate tAMPer to accelerate AMP discovery and development by reducing the reliance on laborious toxicity screening experiments.","PeriodicalId":20761,"journal":{"name":"Protein Science","volume":null,"pages":null},"PeriodicalIF":4.5000,"publicationDate":"2024-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Structure‐aware deep learning model for peptide toxicity prediction\",\"authors\":\"Hossein Ebrahimikondori, Darcy Sutherland, Anat Yanai, Amelia Richter, Ali Salehi, Chenkai Li, Lauren Coombe, Monica Kotkoff, René L. Warren, Inanc Birol\",\"doi\":\"10.1002/pro.5076\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Antimicrobial resistance is a critical public health concern, necessitating the exploration of alternative treatments. While antimicrobial peptides (AMPs) show promise, assessing their toxicity using traditional wet lab methods is both time‐consuming and costly. We introduce tAMPer, a novel multi‐modal deep learning model designed to predict peptide toxicity by integrating the underlying amino acid sequence composition and the three‐dimensional structure of peptides. tAMPer adopts a graph‐based representation for peptides, encoding ColabFold‐predicted structures, where nodes represent amino acids and edges represent spatial interactions. Structural features are extracted using graph neural networks, and recurrent neural networks capture sequential dependencies. tAMPer's performance was assessed on a publicly available protein toxicity benchmark and an AMP hemolysis data we generated. On the latter, tAMPer achieves an F1‐score of 68.7%, outperforming the second‐best method by 23.4%. On the protein benchmark, tAMPer exhibited an improvement of over 3.0% in the F1‐score compared to current state‐of‐the‐art methods. We anticipate tAMPer to accelerate AMP discovery and development by reducing the reliance on laborious toxicity screening experiments.\",\"PeriodicalId\":20761,\"journal\":{\"name\":\"Protein Science\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":4.5000,\"publicationDate\":\"2024-06-22\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Protein Science\",\"FirstCategoryId\":\"99\",\"ListUrlMain\":\"https://doi.org/10.1002/pro.5076\",\"RegionNum\":3,\"RegionCategory\":\"生物学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"BIOCHEMISTRY & MOLECULAR BIOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Protein Science","FirstCategoryId":"99","ListUrlMain":"https://doi.org/10.1002/pro.5076","RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"BIOCHEMISTRY & MOLECULAR BIOLOGY","Score":null,"Total":0}
Structure‐aware deep learning model for peptide toxicity prediction
Antimicrobial resistance is a critical public health concern, necessitating the exploration of alternative treatments. While antimicrobial peptides (AMPs) show promise, assessing their toxicity using traditional wet lab methods is both time‐consuming and costly. We introduce tAMPer, a novel multi‐modal deep learning model designed to predict peptide toxicity by integrating the underlying amino acid sequence composition and the three‐dimensional structure of peptides. tAMPer adopts a graph‐based representation for peptides, encoding ColabFold‐predicted structures, where nodes represent amino acids and edges represent spatial interactions. Structural features are extracted using graph neural networks, and recurrent neural networks capture sequential dependencies. tAMPer's performance was assessed on a publicly available protein toxicity benchmark and an AMP hemolysis data we generated. On the latter, tAMPer achieves an F1‐score of 68.7%, outperforming the second‐best method by 23.4%. On the protein benchmark, tAMPer exhibited an improvement of over 3.0% in the F1‐score compared to current state‐of‐the‐art methods. We anticipate tAMPer to accelerate AMP discovery and development by reducing the reliance on laborious toxicity screening experiments.
期刊介绍:
Protein Science, the flagship journal of The Protein Society, is a publication that focuses on advancing fundamental knowledge in the field of protein molecules. The journal welcomes original reports and review articles that contribute to our understanding of protein function, structure, folding, design, and evolution.
Additionally, Protein Science encourages papers that explore the applications of protein science in various areas such as therapeutics, protein-based biomaterials, bionanotechnology, synthetic biology, and bioelectronics.
The journal accepts manuscript submissions in any suitable format for review, with the requirement of converting the manuscript to journal-style format only upon acceptance for publication.
Protein Science is indexed and abstracted in numerous databases, including the Agricultural & Environmental Science Database (ProQuest), Biological Science Database (ProQuest), CAS: Chemical Abstracts Service (ACS), Embase (Elsevier), Health & Medical Collection (ProQuest), Health Research Premium Collection (ProQuest), Materials Science & Engineering Database (ProQuest), MEDLINE/PubMed (NLM), Natural Science Collection (ProQuest), and SciTech Premium Collection (ProQuest).