Kun Li, Xiuwen Gong, Shirui Pan, Jia Wu, Bo Du, Wenbin Hu
{"title":"支持药物反应预测的无调节因子分子生成","authors":"Kun Li, Xiuwen Gong, Shirui Pan, Jia Wu, Bo Du, Wenbin Hu","doi":"arxiv-2405.14536","DOIUrl":null,"url":null,"abstract":"Drug response prediction (DRP) is a crucial phase in drug discovery, and the\nmost important metric for its evaluation is the IC50 score. DRP results are\nheavily dependent on the quality of the generated molecules. Existing molecule\ngeneration methods typically employ classifier-based guidance, enabling\nsampling within the IC50 classification range. However, these methods fail to\nensure the sampling space range's effectiveness, generating numerous\nineffective molecules. Through experimental and theoretical study, we\nhypothesize that conditional generation based on the target IC50 score can\nobtain a more effective sampling space. As a result, we introduce\nregressor-free guidance molecule generation to ensure sampling within a more\neffective space and support DRP. Regressor-free guidance combines a diffusion\nmodel's score estimation with a regression controller model's gradient based on\nnumber labels. To effectively map regression labels between drugs and cell\nlines, we design a common-sense numerical knowledge graph that constrains the\norder of text representations. Experimental results on the real-world dataset\nfor the DRP task demonstrate our method's effectiveness in drug discovery. The\ncode is available at:https://anonymous.4open.science/r/RMCD-DBD1.","PeriodicalId":501325,"journal":{"name":"arXiv - QuanBio - Molecular Networks","volume":"29 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Regressor-free Molecule Generation to Support Drug Response Prediction\",\"authors\":\"Kun Li, Xiuwen Gong, Shirui Pan, Jia Wu, Bo Du, Wenbin Hu\",\"doi\":\"arxiv-2405.14536\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Drug response prediction (DRP) is a crucial phase in drug discovery, and the\\nmost important metric for its evaluation is the IC50 score. DRP results are\\nheavily dependent on the quality of the generated molecules. Existing molecule\\ngeneration methods typically employ classifier-based guidance, enabling\\nsampling within the IC50 classification range. However, these methods fail to\\nensure the sampling space range's effectiveness, generating numerous\\nineffective molecules. Through experimental and theoretical study, we\\nhypothesize that conditional generation based on the target IC50 score can\\nobtain a more effective sampling space. As a result, we introduce\\nregressor-free guidance molecule generation to ensure sampling within a more\\neffective space and support DRP. Regressor-free guidance combines a diffusion\\nmodel's score estimation with a regression controller model's gradient based on\\nnumber labels. To effectively map regression labels between drugs and cell\\nlines, we design a common-sense numerical knowledge graph that constrains the\\norder of text representations. Experimental results on the real-world dataset\\nfor the DRP task demonstrate our method's effectiveness in drug discovery. The\\ncode is available at:https://anonymous.4open.science/r/RMCD-DBD1.\",\"PeriodicalId\":501325,\"journal\":{\"name\":\"arXiv - QuanBio - Molecular Networks\",\"volume\":\"29 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-05-23\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - QuanBio - Molecular Networks\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2405.14536\",\"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 - Molecular Networks","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2405.14536","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Regressor-free Molecule Generation to Support Drug Response Prediction
Drug response prediction (DRP) is a crucial phase in drug discovery, and the
most important metric for its evaluation is the IC50 score. DRP results are
heavily dependent on the quality of the generated molecules. Existing molecule
generation methods typically employ classifier-based guidance, enabling
sampling within the IC50 classification range. However, these methods fail to
ensure the sampling space range's effectiveness, generating numerous
ineffective molecules. Through experimental and theoretical study, we
hypothesize that conditional generation based on the target IC50 score can
obtain a more effective sampling space. As a result, we introduce
regressor-free guidance molecule generation to ensure sampling within a more
effective space and support DRP. Regressor-free guidance combines a diffusion
model's score estimation with a regression controller model's gradient based on
number labels. To effectively map regression labels between drugs and cell
lines, we design a common-sense numerical knowledge graph that constrains the
order of text representations. Experimental results on the real-world dataset
for the DRP task demonstrate our method's effectiveness in drug discovery. The
code is available at:https://anonymous.4open.science/r/RMCD-DBD1.