Evening the Score: Targeting SARS-CoV-2 Protease Inhibition in Graph Generative Models for Therapeutic Candidates

TL;DR

This paper compares two graph generative models for designing SARS-CoV-2 drug candidates, focusing on balancing druglikeness, synthetic accessibility, and antiviral activity to accelerate pandemic response drug discovery.

Contribution

It introduces a combined framework using autoencoders and reinforcement learning to generate and optimize novel therapeutic molecules targeting SARS-CoV-2.

Findings

Autoencoder generates molecules similar to known anti-SARS drugs.

Reinforcement learning produces highly novel molecules.

Framework enables high-throughput targeted drug candidate generation.

Abstract

We examine a pair of graph generative models for the therapeutic design of novel drug candidates targeting SARS-CoV-2 viral proteins. Due to a sense of urgency, we chose well-validated models with unique strengths: an autoencoder that generates molecules with similar structures to a dataset of drugs with anti-SARS activity and a reinforcement learning algorithm that generates highly novel molecules. During generation, we explore optimization toward several design targets to balance druglikeness, synthetic accessability, and anti-SARS activity based on \icfifty. This generative framework\footnote{https://github.com/exalearn/covid-drug-design} will accelerate drug discovery in future pandemics through the high-throughput generation of targeted therapeutic candidates.