{"title":"探索基于图形的模型预测抗三阴性乳腺癌的活性化合物。","authors":"Hridoy Jyoti Mahanta, Amarjeet Boruah, Bikram Phukan, Hillul Chutia, Pankaj Bharali, Selvaraman Nagamani","doi":"10.1007/s11030-025-11283-7","DOIUrl":null,"url":null,"abstract":"<p><p>Breast cancer is among the most dominant and rapidly rising cancers, both in India and around the world. Triple-negative breast cancer (TNBC) is one of the most aggressive subtypes of breast cancer, distinguished by the absence of HER2, progesterone, and estrogen receptor expressions. This absence limits treatment options, emphasizing the urgent need to discover or design new drug candidates for TNBC. Integrating artificial intelligence and machine learning in computational modeling, has significantly accelerated the analysis of large-scale biological data and improved the prediction of therapeutic outcomes. In this study, we curated a data set of 756 mutant-type compounds from three cell lines and developed four graph-based models to predict active compounds against TNBC. Validated using stratified nested tenfold cross-validation and optimized with the Optuna framework, the models achieved predictive accuracy with AUC values of 0.65-0.82, with the MPNN model outperforming all the others. Furthermore, key structural fragments associated with cell inhibition and model predictions were identified and interpreted using several explainability techniques. Validation with an external set of FDA-approved drugs demonstrated prediction accuracies ranging from 66% to 97%, highlighting the robustness of the models in identifying compounds with potential inhibitory activity against TNBC cells.</p>","PeriodicalId":708,"journal":{"name":"Molecular Diversity","volume":" ","pages":""},"PeriodicalIF":3.8000,"publicationDate":"2025-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Exploring graph-based models for predicting active compounds against triple-negative breast cancer.\",\"authors\":\"Hridoy Jyoti Mahanta, Amarjeet Boruah, Bikram Phukan, Hillul Chutia, Pankaj Bharali, Selvaraman Nagamani\",\"doi\":\"10.1007/s11030-025-11283-7\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>Breast cancer is among the most dominant and rapidly rising cancers, both in India and around the world. Triple-negative breast cancer (TNBC) is one of the most aggressive subtypes of breast cancer, distinguished by the absence of HER2, progesterone, and estrogen receptor expressions. This absence limits treatment options, emphasizing the urgent need to discover or design new drug candidates for TNBC. Integrating artificial intelligence and machine learning in computational modeling, has significantly accelerated the analysis of large-scale biological data and improved the prediction of therapeutic outcomes. In this study, we curated a data set of 756 mutant-type compounds from three cell lines and developed four graph-based models to predict active compounds against TNBC. Validated using stratified nested tenfold cross-validation and optimized with the Optuna framework, the models achieved predictive accuracy with AUC values of 0.65-0.82, with the MPNN model outperforming all the others. Furthermore, key structural fragments associated with cell inhibition and model predictions were identified and interpreted using several explainability techniques. Validation with an external set of FDA-approved drugs demonstrated prediction accuracies ranging from 66% to 97%, highlighting the robustness of the models in identifying compounds with potential inhibitory activity against TNBC cells.</p>\",\"PeriodicalId\":708,\"journal\":{\"name\":\"Molecular Diversity\",\"volume\":\" \",\"pages\":\"\"},\"PeriodicalIF\":3.8000,\"publicationDate\":\"2025-07-09\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Molecular Diversity\",\"FirstCategoryId\":\"92\",\"ListUrlMain\":\"https://doi.org/10.1007/s11030-025-11283-7\",\"RegionNum\":2,\"RegionCategory\":\"化学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"CHEMISTRY, APPLIED\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Molecular Diversity","FirstCategoryId":"92","ListUrlMain":"https://doi.org/10.1007/s11030-025-11283-7","RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"CHEMISTRY, APPLIED","Score":null,"Total":0}
Exploring graph-based models for predicting active compounds against triple-negative breast cancer.
Breast cancer is among the most dominant and rapidly rising cancers, both in India and around the world. Triple-negative breast cancer (TNBC) is one of the most aggressive subtypes of breast cancer, distinguished by the absence of HER2, progesterone, and estrogen receptor expressions. This absence limits treatment options, emphasizing the urgent need to discover or design new drug candidates for TNBC. Integrating artificial intelligence and machine learning in computational modeling, has significantly accelerated the analysis of large-scale biological data and improved the prediction of therapeutic outcomes. In this study, we curated a data set of 756 mutant-type compounds from three cell lines and developed four graph-based models to predict active compounds against TNBC. Validated using stratified nested tenfold cross-validation and optimized with the Optuna framework, the models achieved predictive accuracy with AUC values of 0.65-0.82, with the MPNN model outperforming all the others. Furthermore, key structural fragments associated with cell inhibition and model predictions were identified and interpreted using several explainability techniques. Validation with an external set of FDA-approved drugs demonstrated prediction accuracies ranging from 66% to 97%, highlighting the robustness of the models in identifying compounds with potential inhibitory activity against TNBC cells.
期刊介绍:
Molecular Diversity is a new publication forum for the rapid publication of refereed papers dedicated to describing the development, application and theory of molecular diversity and combinatorial chemistry in basic and applied research and drug discovery. The journal publishes both short and full papers, perspectives, news and reviews dealing with all aspects of the generation of molecular diversity, application of diversity for screening against alternative targets of all types (biological, biophysical, technological), analysis of results obtained and their application in various scientific disciplines/approaches including:
combinatorial chemistry and parallel synthesis;
small molecule libraries;
microwave synthesis;
flow synthesis;
fluorous synthesis;
diversity oriented synthesis (DOS);
nanoreactors;
click chemistry;
multiplex technologies;
fragment- and ligand-based design;
structure/function/SAR;
computational chemistry and molecular design;
chemoinformatics;
screening techniques and screening interfaces;
analytical and purification methods;
robotics, automation and miniaturization;
targeted libraries;
display libraries;
peptides and peptoids;
proteins;
oligonucleotides;
carbohydrates;
natural diversity;
new methods of library formulation and deconvolution;
directed evolution, origin of life and recombination;
search techniques, landscapes, random chemistry and more;