Yijia Xiao, Edward Sun, Yiqiao Jin, Qifan Wang, Wei Wang
{"title":"ProteinGPT:用于蛋白质特性预测和结构理解的多模式 LLM","authors":"Yijia Xiao, Edward Sun, Yiqiao Jin, Qifan Wang, Wei Wang","doi":"arxiv-2408.11363","DOIUrl":null,"url":null,"abstract":"Understanding biological processes, drug development, and biotechnological\nadvancements requires detailed analysis of protein structures and sequences, a\ntask in protein research that is inherently complex and time-consuming when\nperformed manually. To streamline this process, we introduce ProteinGPT, a\nstate-of-the-art multi-modal protein chat system, that allows users to upload\nprotein sequences and/or structures for comprehensive protein analysis and\nresponsive inquiries. ProteinGPT seamlessly integrates protein sequence and\nstructure encoders with linear projection layers for precise representation\nadaptation, coupled with a large language model (LLM) to generate accurate and\ncontextually relevant responses. To train ProteinGPT, we construct a\nlarge-scale dataset of 132,092 proteins with annotations, and optimize the\ninstruction-tuning process using GPT-4o. This innovative system ensures\naccurate alignment between the user-uploaded data and prompts, simplifying\nprotein analysis. Experiments show that ProteinGPT can produce promising\nresponses to proteins and their corresponding questions.","PeriodicalId":501022,"journal":{"name":"arXiv - QuanBio - Biomolecules","volume":"3 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"ProteinGPT: Multimodal LLM for Protein Property Prediction and Structure Understanding\",\"authors\":\"Yijia Xiao, Edward Sun, Yiqiao Jin, Qifan Wang, Wei Wang\",\"doi\":\"arxiv-2408.11363\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Understanding biological processes, drug development, and biotechnological\\nadvancements requires detailed analysis of protein structures and sequences, a\\ntask in protein research that is inherently complex and time-consuming when\\nperformed manually. To streamline this process, we introduce ProteinGPT, a\\nstate-of-the-art multi-modal protein chat system, that allows users to upload\\nprotein sequences and/or structures for comprehensive protein analysis and\\nresponsive inquiries. ProteinGPT seamlessly integrates protein sequence and\\nstructure encoders with linear projection layers for precise representation\\nadaptation, coupled with a large language model (LLM) to generate accurate and\\ncontextually relevant responses. To train ProteinGPT, we construct a\\nlarge-scale dataset of 132,092 proteins with annotations, and optimize the\\ninstruction-tuning process using GPT-4o. This innovative system ensures\\naccurate alignment between the user-uploaded data and prompts, simplifying\\nprotein analysis. Experiments show that ProteinGPT can produce promising\\nresponses to proteins and their corresponding questions.\",\"PeriodicalId\":501022,\"journal\":{\"name\":\"arXiv - QuanBio - Biomolecules\",\"volume\":\"3 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-08-21\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - QuanBio - Biomolecules\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2408.11363\",\"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 - Biomolecules","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2408.11363","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
ProteinGPT: Multimodal LLM for Protein Property Prediction and Structure Understanding
Understanding biological processes, drug development, and biotechnological
advancements requires detailed analysis of protein structures and sequences, a
task in protein research that is inherently complex and time-consuming when
performed manually. To streamline this process, we introduce ProteinGPT, a
state-of-the-art multi-modal protein chat system, that allows users to upload
protein sequences and/or structures for comprehensive protein analysis and
responsive inquiries. ProteinGPT seamlessly integrates protein sequence and
structure encoders with linear projection layers for precise representation
adaptation, coupled with a large language model (LLM) to generate accurate and
contextually relevant responses. To train ProteinGPT, we construct a
large-scale dataset of 132,092 proteins with annotations, and optimize the
instruction-tuning process using GPT-4o. This innovative system ensures
accurate alignment between the user-uploaded data and prompts, simplifying
protein analysis. Experiments show that ProteinGPT can produce promising
responses to proteins and their corresponding questions.