{"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/smiles2vec-an-interpretable-general-purpose","title":"SMILES2Vec: An Interpretable General-Purpose Deep Neural Network for Predicting Chemical Properties","arxiv_id":"1712.02034","date":"2017-12-06","proceeding":null,"authors":["Garrett B. Goh","Nathan O. Hodas","Charles Siegel","Abhinav Vishnu"],"abstract":"Chemical databases store information in text representations, and the SMILES\nformat is a universal standard used in many cheminformatics software. Encoded\nin each SMILES string is structural information that can be used to predict\ncomplex chemical properties. In this work, we develop SMILES2vec, a deep RNN\nthat automatically learns features from SMILES to predict chemical properties,\nwithout the need for additional explicit feature engineering. Using Bayesian\noptimization methods to tune the network architecture, we show that an\noptimized SMILES2vec model can serve as a general-purpose neural network for\npredicting distinct chemical properties including toxicity, activity,\nsolubility and solvation energy, while also outperforming contemporary MLP\nneural networks that uses engineered features. Furthermore, we demonstrate\nproof-of-concept of interpretability by developing an explanation mask that\nlocalizes on the most important characters used in making a prediction. When\ntested on the solubility dataset, it identified specific parts of a chemical\nthat is consistent with established first-principles knowledge with an accuracy\nof 88%. Our work demonstrates that neural networks can learn technically\naccurate chemical concept and provide state-of-the-art accuracy, making\ninterpretable deep neural networks a useful tool of relevance to the chemical\nindustry.","url_abs":"http://arxiv.org/abs/1712.02034v2","url_pdf":"http://arxiv.org/pdf/1712.02034v2.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":"smiles2vec-an-interpretable-general-purpose","repo_url":"https://github.com/Abdulk084/Smiles2vec","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"smiles2vec-an-interpretable-general-purpose","repo_url":"https://github.com/Abhik1998/HIV-predictor","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"smiles2vec-an-interpretable-general-purpose","repo_url":"https://github.com/liambll/drug-activity-prediction","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"smiles2vec-an-interpretable-general-purpose","repo_url":"https://github.com/liambll/drug-efficacy-prediction","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"bayesian-optimization","task_name":"Bayesian Optimization"},{"task_slug":"feature-engineering","task_name":"Feature Engineering"}],"methods":[{"method_slug":"interpretability","method_name":"Interpretability"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1712.02034","atlas_url":"https://app.syntology.ai/?focus=1712.02034","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}