[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-290401-105":59,"doc-detail-290401-en":130},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","the-use-of-machine-learning-in-human-cytochrome-p450s-substrate-prediction","The Use of Machine Learning in Human Cytochrome P450s Substrate Prediction","","Cytochrome P450 enzymes (CYP450s) play a central role in human drug oxidation, and predicting whether chemicals act as substrates is important for drug discovery and toxicology screening. The document surveys structure-based and ligand-based computational strategies, emphasizing QSAR and the growing role of machine learning for QSAR modeling. It evaluates multiple published in silico CYP450 prediction tools and concludes that classical ML and deep learning can both reach useful accuracy, while robustness and performance remain improvement targets.",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/the-use-of-machine-learning-in-human-cytochrome-p450s-substrate-prediction/290401/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/the-use-of-machine-learning-in-human-cytochrome-p450s-substrate-prediction/290401.png","ImageObject",300,407,{"name":92,"@type":93},"Eliana","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-18","2026-09-17",true,{"@type":102,"interactionType":103,"userInteractionCount":44},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What computational approaches are used to predict chemical interactions with human CYP450s?","Question",{"text":112,"@type":113},"The document describes structure-based approaches that use 3D protein structures and ligand-based approaches that use similarity and QSAR modeling of molecular descriptors. It notes that machine learning now commonly underpins QSAR model building.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"How does the CypEGAT model represent proteins and molecules?",{"text":117,"@type":113},"CypEGAT uses a pre-trained ESM-2 Transformer to encode CYP450s protein representations and a fine-tuned molecular graph attention network to generate molecular representations. An XGBoost classifier then determines substrate versus non-substrate status for each CYP450 isoform.",{"name":119,"@type":110,"acceptedAnswer":120},"What are the main takeaways about ML vs deep learning performance from the document?",{"text":121,"@type":113},"The document states that classical machine learning and deep learning can both achieve prediction accuracy. It also highlights that classical ML is constrained by dataset scope and computational resources, while deep learning models are newer and still need improved robustness and performance.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},290401,1789641615,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":44,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":8,"language":139,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":140,"faqs":141,"seo_title":142,"seo_description":67,"update_tm":129,"read_time":81,"preview_page_count":4},4398048949847,"https://ap-avatar.wpscdn.com/avatar/400002536579ef2da7f?_k=1778318612642679267","The Use of Machine Learning in Human Cytochrome P450s Substrate Prediction  \nYao Wei 1 , Uliano Guerrini 1, Ivano Eberini 1  \n1 Department of Pharmacological and Biomolecular Sciences \"Rodolfo Paoletti\", DiSFeB, University of Milan, Via Balzaretti 9 20133 Milan, Italy  \n[yao.wei@unimi.it](yao.wei@unimi.it)  \nIntroduction  \nCytochrome P450s (CYP450s), a highly diverse superfamily of heme-thiolate proteins, are indispensable components of the oxidative metabolic machinery found across various life forms . In humans, 57 distinct CYP450 isoforms have been identified, collectively responsible for metabolizing 70-80% of clinically used drugs . Computational approaches to predict interactions between chemical compounds and CYP450s offer significant advantages, such as reducing economic and labor costs, alleviating environmental pollution, and facilitating the preselection of hit compounds for drug discovery and toxicology studies, thus accelerating research progress .  \nComputational techniques are based on two main approaches: structure-based and ligandbased approaches . The former one relies on the available three-dimensional (3D) protein structures to directly assess the interactions between CYP450s and chemicals. On the other hand, ligand-based approaches can evaluate the structural similarities between ligands and known substrates . The predominant ligand-based method is the QSAR modeling which is used to establish correlations between molecular descriptors of chemicals and their biological activities through mathematical models, thereby explaining the inherent relationships at a molecular level. Nowadays, machine learning algorithms have become the mainstay for building the QSAR models .  \nEvaluation of Published In Silico CYP450s Prediction Tools  \nTesting 7 machine learning prediction models performance  \n| Model | Methods | Year |\n| --- | --- | --- |\n| pkCSM | RF, LR | 2015 |\n| CypReact | Learning-based model | 2018 |\n| admetSAR 2.0 | RF, k-NN, SVM | 2019 |\n| CYPstrate | RF, SVM | 2021 |\n| ADMETlab 2.0 | multi-task graph attention | 2021 |\n| ESP | Esm-1b transformer | 2023 |\n| ADMETlab 3.0 | Multi-task DMPNN | 2024 |\n\nConclusion:  \n• Both classical machine learning methods and deep learning methods can achieve a certain prediction accuracy.  \n• Classical machine learning-based models are limited by their datasets scope and computational resources .  \n• Deep learning-based models are newly developed and models robustness and performance still need to be improved.  \nReference:  \n[1] Wei, Yao, et al. \"Investigation of in silico studies for cytochrome P450 isoforms specificity. \" Computational and Structural Biotechnology Journal (2024) .  \n[2] Wei, Yao, et al. \"CypEGAT: A Deep Learning Framework Integrating Protein Language Model and Graph Attention Networks for Enhanced CYP450s Substrate Prediction . \" AAAI2025, AI4Research workshop.  \nFunding:  \nThis work was supported by the European Union’s Horizon Europe research and innovation program under the Marie Skłodowska-Curie grant agreement No. 101073546 (MSCA Doctoral Network Metal-containing Radical Enzymes – MetRaZymes), and grants from MIUR- “Progetto Eccellenza 2023 – 2027 \" .  \nOur CYP450s Substrate Prediction Machine Learning Model: CypEGAT We present an enhanced deep learning framework, CypEGAT, designed to predict substrates for five key human CYP450 isoforms . The model incorporates the pre-trained ESM-2 Transformer to encode CYP450s protein representations and employs a fine-tuned molecular graph attention network to generate molecular representations . In the classification module, we utilize an XGBoost classifier to determine whether a given compound is a substrate or non-substrate for a specific CYP450 isoform .  \nOur Fine-tuned Molecular Graph Attention Network (GAT)  \nOur molecular GAT employs multi-head attention, adjacency masking, and dropout for robust learning and predicts final outputs using fully connected layers . It consists of two GAT layers: the first l","cbCaivJxL7ULZfle","https://ap.wps.com/l/cbCaivJxL7ULZfle","pdf",10815413,"English","# Introduction\n# Evaluation of Published In Silico CYP450s Prediction Tools\n# Conclusion\n# Our CYP450s Substrate Prediction Machine Learning Model: CypEGAT\n## Our Fine-tuned Molecular Graph Attention Network (GAT)\n# Results\n# Reference\n# Funding","[{\"question\":\"What computational approaches are used to predict chemical interactions with human CYP450s?\",\"answer\":\"The document describes structure-based approaches that use 3D protein structures and ligand-based approaches that use similarity and QSAR modeling of molecular descriptors. It notes that machine learning now commonly underpins QSAR model building.\"},{\"question\":\"How does the CypEGAT model represent proteins and molecules?\",\"answer\":\"CypEGAT uses a pre-trained ESM-2 Transformer to encode CYP450s protein representations and a fine-tuned molecular graph attention network to generate molecular representations. An XGBoost classifier then determines substrate versus non-substrate status for each CYP450 isoform.\"},{\"question\":\"What are the main takeaways about ML vs deep learning performance from the document?\",\"answer\":\"The document states that classical machine learning and deep learning can both achieve prediction accuracy. It also highlights that classical ML is constrained by dataset scope and computational resources, while deep learning models are newer and still need improved robustness and performance.\"}]","The Use of Machine Learning in Human Cytochrome P450s Substrate Prediction | PDF"]