{"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/interpretable-deep-learning-in-drug-discovery","title":"Interpretable Deep Learning in Drug Discovery","arxiv_id":"1903.02788","date":"2019-03-07","proceeding":null,"authors":["Kristina Preuer","Günter Klambauer","Friedrich Rippmann","Sepp Hochreiter","Thomas Unterthiner"],"abstract":"Without any means of interpretation, neural networks that predict molecular\nproperties and bioactivities are merely black boxes. We will unravel these\nblack boxes and will demonstrate approaches to understand the learned\nrepresentations which are hidden inside these models. We show how single\nneurons can be interpreted as classifiers which determine the presence or\nabsence of pharmacophore- or toxicophore-like structures, thereby generating\nnew insights and relevant knowledge for chemistry, pharmacology and\nbiochemistry. We further discuss how these novel pharmacophores/toxicophores\ncan be determined from the network by identifying the most relevant components\nof a compound for the prediction of the network. Additionally, we propose a\nmethod which can be used to extract new pharmacophores from a model and will\nshow that these extracted structures are consistent with literature findings.\nWe envision that having access to such interpretable knowledge is a crucial aid\nin the development and design of new pharmaceutically active molecules, and\nhelps to investigate and understand failures and successes of current methods.","url_abs":"http://arxiv.org/abs/1903.02788v2","url_pdf":"http://arxiv.org/pdf/1903.02788v2.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":"interpretable-deep-learning-in-drug-discovery","repo_url":"https://github.com/bioinf-jku/interpretable_ml_drug_discovery","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"deep-learning","task_name":"Deep Learning"},{"task_slug":"drug-discovery","task_name":"Drug Discovery"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1903.02788","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}