Semi-equivariant conditional normalizing flows, with applications to target-aware molecule generation

Abstract

Learning over the domain of 3D graphs has applications in a number of scientific and engineering disciplines, including molecular chemistry, high energy physics, and computer vision. We consider a specific problem in this domain, namely: given one such 3D graph, dubbed the base graph, our goal is to learn a conditional distribution over another such graph, dubbed the complement graph. Due to the three-dimensional nature of the graphs in question, there are certain natural invariances such a distribution should satisfy: it should be invariant to rigid body transformations that act jointly on the base graph and the complement graph, and it should also be invariant to permutations of the vertices of either graph. We propose a general method for learning the conditional probabilistic model, the central part of which is a continuous normalizing flow. We establish semi-equivariance conditions on the flow which guarantee the aforementioned invariance conditions on the conditional distribution. Additionally, we propose a graph neural network architecture which implements this flow, and which is designed to learn effectively despite the typical differences in size between the base graph and the complement graph. We demonstrate the utility of our technique in the molecular setting by training a conditional generative model which, given a receptor, can generate ligands which may successfully bind to that receptor. The resulting model, which has potential applications in drug design, displays high quality performance in the key ∆Binding metric.

A Review on the Recent Applications of Deep Learning in Predictive Drug Toxicological Studies

“…In contrast, the conditional setting involves generating ligand molecules to bind to a target receptor, which is more applicable to drug design. Rozenberg et al present a method for generating ligand molecules that can bind to a given receptor molecule based on a conditional generative model . In other words, this conditional generative model generates target-specific ligands that have the potential to lessen the risk of nonspecific ADRs.…”

“…In other words, this conditional generative model generates target-specific ligands that have the potential to lessen the risk of nonspecific ADRs. The algorithm uses a continuous normalizing flow to learn a distribution that is invariant to rigid body transformations and permutations of the ligand and receptor atoms . The flow is implemented using a GNN architecture that can handle large differences in size between the ligand and receptor .…”

“…The algorithm uses a continuous normalizing flow to learn a distribution that is invariant to rigid body transformations and permutations of the ligand and receptor atoms . The flow is implemented using a GNN architecture that can handle large differences in size between the ligand and receptor . Authors tested their approach on the CrossDocked2020 data set, achieving better results in binding affinity than with other methods …”

“…The flow is implemented using a GNN architecture that can handle large differences in size between the ligand and receptor . Authors tested their approach on the CrossDocked2020 data set, achieving better results in binding affinity than with other methods …”