During the lead optimization phase of the drug discovery pipeline, lead molecules are refined for potency via slight modifications of their chemical structure. Relative binding free energy (RBFE) methods allow comparisons of molecular potency during this optimization. We utilize a Siamese Convolutional Neural Network (CNN) to directly estimate the RBFE with higher throughput than simulation based methods. Our models show improved performance over a previously published Siamese RBFE predictor. The components of the models' architecture and training objective are investigated for their effect on performance. We explore methods to boost poor, out-of-domain, RBFE prediction performance with limited additional training data.

Exploring relative binding free energy prediction with Siamese networks

ACS Spring 2022

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.