Abstract

The Programmed Cell Death Protein 1/Programmed Death-Ligand 1 (PD-1/PD-L1) interaction is an immune checkpoint utilized by cancer cells to enhance immune suppression. There is a huge need to develop small molecule drugs that are fast acting, cost effective, and readily bioavailable compared to antibodies. Unfortunately, synthesizing and validating large libraries of small- molecules to inhibit PD-1/PD-L1 interaction in a blind manner is both time-consuming and expensive. To improve this drug discovery pipeline, we have developed a machine learning methodology trained on patent data to identify, synthesize, and validate PD-1/PD-L1 small molecule inhibitors. Our model incorporates two features: docking scores to represent the energy of binding (E) as a global feature and sub-graph features through a graph neural network (GNN) of molecular topology to represent local features. This interaction energy-based Graph Neural Network (EGNN) model outperforms traditional machine learning methods and a simple GNN with a F1 score of 0.9524 and Cohen’s kappa score of 0.8861 for the hold out test set, suggesting that the topology of the small molecule, the structural interaction in the binding pocket, and chemical diversity of the training data are all important considerations for enhancing model performance. A Bootstrapped EGNN model was used to select compounds for synthesis and experimental validation with predicted high and low potency to inhibit PD-1/PD-L1 interaction. The potent inhibitor, (4-((3-(2,3-dihydrobenzo[b][1,4]dioxin-6-yl)-2- methylbenzyl)oxy)-2,6-dimethoxybenzyl)-D-serine, is a hybrid of two known bioactive scaffolds, with an IC50 of 339.9 nM that is comparatively better than the known bioactive compound. We conclude that our bootstrapped EGNN model will be useful to identify target-specific high potency molecules designed by scaffold hopping, a well-known medicinal chemistry technique.