Artificial Intelligence (AI) based models have demonstrated their ability to explore the huge chemical space efficiently and to find out novel candidates with target activity. The molecule generation and lead optimization while preserving critical scaffolds is an efficient and most effective way to achieve not only desired activity but also efficacy. Existing approach currently focus mainly on generating molecules in the form of SMILES strings only. We propose a novel generative model that can take 3D geometry as an input and generates 3D coordinates of novel candidates as an output while always preserving the desired scaffolds. Availability of co-crystal structures for main protease (Mpro) and potential inhibitors provides us the key scaffolds essential for binding and inhibition of Mpro to improve the efficacy. We demonstrate that our model generates predominantly valid, unique, novel molecules and have drug-like properties similar to the experimentally validated structures. Moreover, our model performs well for relatively small volume of training data and generalize equally well for generating the molecules with any new scaffolds thus making our model applicable in different domains.

Deep learning to generate 3D-coordinates of drug candidates with desired scaffolds

Where thousands of chemistry professionals meet to share ideas and advance scientific and technical knowledge. Spring 2023 abstracts, posters, and presentations will launch at the end of March.

ACS Spring 2023

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