Aleix Lafita, Ferran Gonzalez, Mahmoud Hossam, Paul Smyth, Jacob Deasy, Ari Allyn-Feuer, Daniel Seaton, Stephen Young
{"title":"利用深度突变扫描微调蛋白质语言模型,提高变异效应预测能力","authors":"Aleix Lafita, Ferran Gonzalez, Mahmoud Hossam, Paul Smyth, Jacob Deasy, Ari Allyn-Feuer, Daniel Seaton, Stephen Young","doi":"arxiv-2405.06729","DOIUrl":null,"url":null,"abstract":"Protein Language Models (PLMs) have emerged as performant and scalable tools\nfor predicting the functional impact and clinical significance of\nprotein-coding variants, but they still lag experimental accuracy. Here, we\npresent a novel fine-tuning approach to improve the performance of PLMs with\nexperimental maps of variant effects from Deep Mutational Scanning (DMS) assays\nusing a Normalised Log-odds Ratio (NLR) head. We find consistent improvements\nin a held-out protein test set, and on independent DMS and clinical variant\nannotation benchmarks from ProteinGym and ClinVar. These findings demonstrate\nthat DMS is a promising source of sequence diversity and supervised training\ndata for improving the performance of PLMs for variant effect prediction.","PeriodicalId":501070,"journal":{"name":"arXiv - QuanBio - Genomics","volume":"10 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Fine-tuning Protein Language Models with Deep Mutational Scanning improves Variant Effect Prediction\",\"authors\":\"Aleix Lafita, Ferran Gonzalez, Mahmoud Hossam, Paul Smyth, Jacob Deasy, Ari Allyn-Feuer, Daniel Seaton, Stephen Young\",\"doi\":\"arxiv-2405.06729\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Protein Language Models (PLMs) have emerged as performant and scalable tools\\nfor predicting the functional impact and clinical significance of\\nprotein-coding variants, but they still lag experimental accuracy. Here, we\\npresent a novel fine-tuning approach to improve the performance of PLMs with\\nexperimental maps of variant effects from Deep Mutational Scanning (DMS) assays\\nusing a Normalised Log-odds Ratio (NLR) head. We find consistent improvements\\nin a held-out protein test set, and on independent DMS and clinical variant\\nannotation benchmarks from ProteinGym and ClinVar. These findings demonstrate\\nthat DMS is a promising source of sequence diversity and supervised training\\ndata for improving the performance of PLMs for variant effect prediction.\",\"PeriodicalId\":501070,\"journal\":{\"name\":\"arXiv - QuanBio - Genomics\",\"volume\":\"10 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-05-10\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - QuanBio - Genomics\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2405.06729\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - QuanBio - Genomics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2405.06729","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Fine-tuning Protein Language Models with Deep Mutational Scanning improves Variant Effect Prediction
Protein Language Models (PLMs) have emerged as performant and scalable tools
for predicting the functional impact and clinical significance of
protein-coding variants, but they still lag experimental accuracy. Here, we
present a novel fine-tuning approach to improve the performance of PLMs with
experimental maps of variant effects from Deep Mutational Scanning (DMS) assays
using a Normalised Log-odds Ratio (NLR) head. We find consistent improvements
in a held-out protein test set, and on independent DMS and clinical variant
annotation benchmarks from ProteinGym and ClinVar. These findings demonstrate
that DMS is a promising source of sequence diversity and supervised training
data for improving the performance of PLMs for variant effect prediction.