Energy-based Generative Models for Target-specific Drug Discovery

TL;DR

This paper introduces TagMol, an energy-based probabilistic model for target-specific drug discovery, demonstrating its ability to generate molecules with comparable binding affinity scores and improved learning efficiency over baseline models.

Contribution

The paper presents a novel energy-based generative model, TagMol, specifically designed for target-specific drug discovery, advancing the capabilities of molecular generation methods.

Findings

TagMol generates molecules with similar binding affinity scores to real molecules.

GAT-based models outperform GCN baselines in learning speed and quality.

The approach enhances computational drug discovery for specific targets.

Abstract

Drug targets are the main focus of drug discovery due to their key role in disease pathogenesis. Computational approaches are widely applied to drug development because of the increasing availability of biological molecular datasets. Popular generative approaches can create new drug molecules by learning the given molecule distributions. However, these approaches are mostly not for target-specific drug discovery. We developed an energy-based probabilistic model for computational target-specific drug discovery. Results show that our proposed TagMol can generate molecules with similar binding affinity scores as real molecules. GAT-based models showed faster and better learning relative to GCN baseline models.