Evaluating Point-Prediction Uncertainties in Neural Networks for Drug Discovery

no code implementations • 31 Oct 2022 • Ya Ju Fan, Jonathan E. Allen, Kevin S. McLoughlin, Da Shi, Brian J. Bennion, Xiaohua Zhang, Felice C. Lightstone

In this paper, we examine UQ methods that estimate different sources of predictive uncertainty for NN models aiming at drug discovery.

Drug Discovery Uncertainty Quantification

High-Throughput Virtual Screening of Small Molecule Inhibitors for SARS-CoV-2 Protein Targets with Deep Fusion Models

no code implementations • 9 Apr 2021 • Garrett A. Stevenson, Derek Jones, Hyojin Kim, W. F. Drew Bennett, Brian J. Bennion, Monica Borucki, Feliza Bourguet, Aidan Epstein, Magdalena Franco, Brooke Harmon, Stewart He, Max P. Katz, Daniel Kirshner, Victoria Lao, Edmond Y. Lau, Jacky Lo, Kevin McLoughlin, Richard Mosesso, Deepa K. Murugesh, Oscar A. Negrete, Edwin A. Saada, Brent Segelke, Maxwell Stefan, Marisa W. Torres, Dina Weilhammer, Sergio Wong, Yue Yang, Adam Zemla, Xiaohua Zhang, Fangqiang Zhu, Felice C. Lightstone, Jonathan E. Allen

Structure-based Deep Fusion models were recently shown to outperform several physics- and machine learning-based protein-ligand binding affinity prediction methods.

BIG-bench Machine Learning