A machine learning model trained on a high-throughput antibacterial screen increases the hit rate of drug discovery
Fig 1
Initial training and performance evaluation of the machine learning model.
(A) High-throughput screening data generated by screening a compound library of 29,537 compounds against B. cenocepacia K56-2 wild-type. Using B-score ≤ -17.5 as a threshold, the screening yielded 256 active compounds. Dark blue and red represent inactive and active compounds, respectively. (B) The machine learning model was trained using a D-MPNN approach, which extracts compounds’ local features, such as atom and bond features. The model was fed more than 200 additional global molecular descriptors to further increase the accuracy. Dataset was split into 80:10:10 ratio to train, validate and test the model. (C) ROC-AUC plot evaluating model performance after training. The model attained a ROC-AUC of 0.823. Parts of panel B are modified from Yang et al. [11]. Fig 1B was created with https://biorender.com/.