Improving ΔΔG predictions with a multi-task convolutional Siamese Network

The lead optimization phase of drug discovery refines an initial hit molecule for desired properties, especially potency. Synthesis and experimental testing of the small perturbations during this refinement can be quite costly and time consuming. Relative binding free energy (RBFE, also referred to as ∆∆G) methods allow the estimation of binding free energy changes after small changes to a ligand scaffold. Here we propose and evaluate a Convolutional Neural Network (CNN) Siamese network for the prediction of RBFE between two bound ligands. We show that our multi-task loss is able to improve on a previous state-of-the-art Siamese network for RBFE prediction via increased regularization of the latent space. The Siamese network architecture is well suited to the prediction of RBFE in comparison to a standard CNN trained on the same data (Pearson’s R of 0.553 and 0.5, respectively). When evaluated on a left-out protein family, our CNN Siamese network shows variability in its RBFE predictive performance depending on the protein family being evaluated (Pearson’s R ranging from-0.44 to 0.97). RBFE prediction performance can be improved during generalization by injecting only a few examples (few-shot learning) from the evaluation dataset during model training.