Torsional-GFN: a conditional conformation generator for small molecules

TL;DR

Torsional-GFN is a novel conditional GFlowNet model that efficiently samples molecular conformations proportional to their Boltzmann distribution, enabling zero-shot generalization and promising scalability for drug discovery applications.

Contribution

It introduces Torsional-GFN, a conditional GFlowNet that samples conformations based on a reward function, with zero-shot generalization to unseen molecules and local structures.

Findings

Samples conformations proportional to Boltzmann distribution

Achieves zero-shot generalization to unseen bond lengths and angles

Works effectively across multiple molecules with a single model

Abstract

Generating stable molecular conformations is crucial in several drug discovery applications, such as estimating the binding affinity of a molecule to a target. Recently, generative machine learning methods have emerged as a promising, more efficient method than molecular dynamics for sampling of conformations from the Boltzmann distribution. In this paper, we introduce Torsional-GFN, a conditional GFlowNet specifically designed to sample conformations of molecules proportionally to their Boltzmann distribution, using only a reward function as training signal. Conditioned on a molecular graph and its local structure (bond lengths and angles), Torsional-GFN samples rotations of its torsion angles. Our results demonstrate that Torsional-GFN is able to sample conformations approximately proportional to the Boltzmann distribution for multiple molecules with a single model, and allows for zero-shot generalization to unseen bond lengths and angles coming from the MD simulations for such molecules. Our work presents a promising avenue for scaling the proposed approach to larger molecular systems, achieving zero-shot generalization to unseen molecules, and including the generation of the local structure into the GFlowNet model.