{"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/molecular-graph-convolutions-moving-beyond","title":"Molecular Graph Convolutions: Moving Beyond Fingerprints","arxiv_id":"1603.00856","date":"2016-03-02","proceeding":null,"authors":["Steven Kearnes","Kevin McCloskey","Marc Berndl","Vijay Pande","Patrick Riley"],"abstract":"Molecular \"fingerprints\" encoding structural information are the workhorse of\ncheminformatics and machine learning in drug discovery applications. However,\nfingerprint representations necessarily emphasize particular aspects of the\nmolecular structure while ignoring others, rather than allowing the model to\nmake data-driven decisions. We describe molecular \"graph convolutions\", a\nmachine learning architecture for learning from undirected graphs, specifically\nsmall molecules. Graph convolutions use a simple encoding of the molecular\ngraph---atoms, bonds, distances, etc.---which allows the model to take greater\nadvantage of information in the graph structure. Although graph convolutions do\nnot outperform all fingerprint-based methods, they (along with other\ngraph-based methods) represent a new paradigm in ligand-based virtual screening\nwith exciting opportunities for future improvement.","url_abs":"http://arxiv.org/abs/1603.00856v3","url_pdf":"http://arxiv.org/pdf/1603.00856v3.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":"molecular-graph-convolutions-moving-beyond","repo_url":"https://github.com/VEK239/StructGNN-lipophilicity","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"molecular-graph-convolutions-moving-beyond","repo_url":"https://github.com/susanzhang233/mollykill_2.0","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"machine-learning","task_name":"BIG-bench Machine Learning"},{"task_slug":"drug-discovery","task_name":"Drug Discovery"},{"task_slug":"graph-regression","task_name":"Graph Regression"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/drug-discovery-on-qm9","task":"Drug Discovery","dataset":"QM9","model":"Molecular Graph Convolutions","rank_in_archive_order":11,"of":11,"metrics":{"Error ratio":"2.59"},"uses_additional_data":false},{"leaderboard":"/sota/graph-regression-on-lipophilicity","task":"Graph Regression","dataset":"Lipophilicity","model":"Weave","rank_in_archive_order":12,"of":23,"metrics":{"RMSE":"0.715"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1603.00856","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}