We describe a novel deep learning method for data imputation that accepts both molecular descriptors and sparse experimental data as inputs to exploit the correlations between experimentally measured endpoints, as well as structure-activity relationships (SAR). The method can robustly estimate the confidence in each prediction and improves the accuracy of prediction over conventional quantitative SAR models. We describe practical applications to drug discovery, including pharma-scale collections comprised of over one million compounds and smaller, project-specific data sets. We illustrate how imputation of missing data, combined with the ability to focus on the most confident predictions, guides the selection of compounds and prioritization of experimental resources in hit-to-lead and lead optimization

Practical applications of deep learning to imputation of drug discovery data

2019 Fall National Meeting

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.