Multi-domain Distribution Learning for De Novo Drug Design

TL;DR

DrugFlow is a novel generative model for structure-based drug design that combines flow matching and Markov bridges, enabling accurate, uncertainty-aware sampling of protein-ligand conformations with improved distribution learning.

Contribution

The paper introduces DrugFlow, integrating continuous flow matching with Markov bridges, and extends it to explore protein conformations, achieving state-of-the-art performance in multi-domain drug design.

Findings

State-of-the-art performance in chemical, geometric, and physical data learning.

Effective uncertainty estimation for out-of-distribution detection.

Enhanced sampling of protein conformational landscapes.

Abstract

We introduce DrugFlow, a generative model for structure-based drug design that integrates continuous flow matching with discrete Markov bridges, demonstrating state-of-the-art performance in learning chemical, geometric, and physical aspects of three-dimensional protein-ligand data. We endow DrugFlow with an uncertainty estimate that is able to detect out-of-distribution samples. To further enhance the sampling process towards distribution regions with desirable metric values, we propose a joint preference alignment scheme applicable to both flow matching and Markov bridge frameworks. Furthermore, we extend our model to also explore the conformational landscape of the protein by jointly sampling side chain angles and molecules.