A machine learning model trained on a high-throughput antibacterial screen increases the hit rate of drug discovery

• Published in: PLoS computational biology Vol. 18; no. 10; p. e1010613
• Main Authors: Rahman, A. S. M. Zisanur, Liu, Chengyou, Sturm, Hunter, Hogan, Andrew M, Davis, Rebecca, Hu, Pingzhao, Cardona, Silvia T
• Format: Journal Article
• Language: English
• Published: San Francisco Public Library of Science 13.10.2022; Public Library of Science (PLoS)
• Subjects: Analysis; Antibacterial activity; Antibacterial agents; Antibacterial materials; Antibiotic resistance; Antibiotics; Artificial intelligence; Bacteria; Bacterial infections; Biology and Life Sciences; Classification; Computer and Information Sciences; Datasets; Deep learning; Drug discovery; Drug resistance; Drug resistance in microorganisms; E coli; FDA approval; Identification and classification; Infections; Learning algorithms; Libraries; Machine learning; Medicine and Health Sciences; Message passing; Methods; Natural products; Neural networks; Optimization; Pathogens; Predictions; Properties; Research and Analysis Methods; Screening; Virtual libraries; China; Taiwan
• ISSN: 1553-7358; 1553-734X; 1553-7358
• DOI: 10.1371/journal.pcbi.1010613

Screening for novel antibacterial compounds in small molecule libraries has a low success rate. We applied machine learning (ML)-based virtual screening for antibacterial activity and evaluated its predictive power by experimental validation. We first binarized 29,537 compounds according to their growth inhibitory activity (hit rate 0.87%) against the antibiotic-resistant bacterium Burkholderia cenocepacia and described their molecular features with a directed-message passing neural network (D-MPNN). Then, we used the data to train an ML model that achieved a receiver operating characteristic (ROC) score of 0.823 on the test set. Finally, we predicted antibacterial activity in virtual libraries corresponding to 1,614 compounds from the Food and Drug Administration (FDA)-approved list and 224,205 natural products. Hit rates of 26% and 12%, respectively, were obtained when we tested the top-ranked predicted compounds for growth inhibitory activity against B. cenocepacia, which represents at least a 14-fold increase from the previous hit rate. In addition, more than 51% of the predicted antibacterial natural compounds inhibited ESKAPE pathogens showing that predictions expand beyond the organism-specific dataset to a broad range of bacteria. Overall, the developed ML approach can be used for compound prioritization before screening, increasing the typical hit rate of drug discovery.