{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/modeling-polypharmacy-side-effects-with-graph","title":"Modeling polypharmacy side effects with graph convolutional networks","arxiv_id":"1802.00543","date":"2018-02-02","proceeding":null,"authors":["Marinka Zitnik","Monica Agrawal","Jure Leskovec"],"abstract":"The use of drug combinations, termed polypharmacy, is common to treat\npatients with complex diseases and co-existing conditions. However, a major\nconsequence of polypharmacy is a much higher risk of adverse side effects for\nthe patient. Polypharmacy side effects emerge because of drug-drug\ninteractions, in which activity of one drug may change if taken with another\ndrug. The knowledge of drug interactions is limited because these complex\nrelationships are rare, and are usually not observed in relatively small\nclinical testing. Discovering polypharmacy side effects thus remains an\nimportant challenge with significant implications for patient mortality. Here,\nwe present Decagon, an approach for modeling polypharmacy side effects. The\napproach constructs a multimodal graph of protein-protein interactions,\ndrug-protein target interactions, and the polypharmacy side effects, which are\nrepresented as drug-drug interactions, where each side effect is an edge of a\ndifferent type. Decagon is developed specifically to handle such multimodal\ngraphs with a large number of edge types. Our approach develops a new graph\nconvolutional neural network for multirelational link prediction in multimodal\nnetworks. Decagon predicts the exact side effect, if any, through which a given\ndrug combination manifests clinically. Decagon accurately predicts polypharmacy\nside effects, outperforming baselines by up to 69%. We find that it\nautomatically learns representations of side effects indicative of\nco-occurrence of polypharmacy in patients. Furthermore, Decagon models\nparticularly well side effects with a strong molecular basis, while on\npredominantly non-molecular side effects, it achieves good performance because\nof effective sharing of model parameters across edge types. Decagon creates\nopportunities to use large pharmacogenomic and patient data to flag and\nprioritize side effects for follow-up analysis.","url_abs":"http://arxiv.org/abs/1802.00543v2","url_pdf":"http://arxiv.org/pdf/1802.00543v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"modeling-polypharmacy-side-effects-with-graph","repo_url":"https://github.com/mims-harvard/decagon","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"link-prediction","task_name":"Link Prediction"}],"methods":[{"method_slug":"rgcn","method_name":"RGCN"}],"datasets_introduced":[{"slug":"decagon","name":"Decagon","full_name":"Bio-decagon"}],"methods_introduced":[],"results":[{"leaderboard":"/sota/link-prediction-on-decagon","task":"Link Prediction","dataset":"Decagon","model":"Decagon","rank_in_archive_order":1,"of":2,"metrics":{"AUPRC":"0.832","AUROC":"0.872","mAP@50":"0.803"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1802.00543","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}