TacoGFN: Target-conditioned GFlowNet for Structure-based Drug Design

TL;DR

TacoGFN is a novel GFlowNet-based model that generates protein-binding molecules conditioned on pocket structures, achieving state-of-the-art success rates and binding affinity scores in structure-based drug design.

Contribution

It introduces a reward-conditioned generative approach that overcomes data limitations and outperforms existing methods in structure-based drug design.

Findings

Achieves 56.0% success rate on CrossDocked2020 benchmark.

Median Vina Dock score of -8.44 kcal/mol, outperforming previous models.

Fine-tuning improves success rate to 88.8% and score to -10.93 kcal/mol.

Abstract

Searching the vast chemical space for drug-like molecules that bind with a protein pocket is a challenging task in drug discovery. Recently, structure-based generative models have been introduced which promise to be more efficient by learning to generate molecules for any given protein structure. However, since they learn the distribution of a limited protein-ligand complex dataset, structure-based methods do not yet outperform optimization-based methods that generate binding molecules for just one pocket. To overcome limitations on data while leveraging learning across protein targets, we choose to model the reward distribution conditioned on pocket structure, instead of the training data distribution. We design TacoGFN, a novel GFlowNet-based approach for structure-based drug design, which can generate molecules conditioned on any protein pocket structure with probabilities proportional to its affinity and property rewards. In the generative setting for CrossDocked2020 benchmark, TacoGFN attains a state-of-the-art success rate of $56.0\%$ and $-8.44$ kcal/mol in median Vina Dock score while improving the generation time by multiple orders of magnitude. Fine-tuning TacoGFN further improves the median Vina Dock score to $-10.93$ kcal/mol and the success rate to $88.8\%$, outperforming all optimization-based methods.