Decoding the Protein-ligand Interactions Using Parallel Graph Neural Networks

Abstract

Protein-ligand interactions (PLIs) are fundamental to biochemical research and their identification is crucial for estimating biophysical and biochemical properties for rational therapeutic design. Currently, experimental characterization of these properties is the most accurate method, however, this is very time-consuming and labor-intensive.A number of computational methods have been developed in this context but most of the existing PLI prediction heavily depend on 2D protein sequence data. Here, we present a novel parallel graph neural network (GNN) to integrate knowledge representation and reasoning for PLI prediction to perform deep learning guided by expert knowledge and informed by 3D structural data. We develop two distinct GNN architectures: GNN F is the base implementation that employs distinct featurization to enhance domain-awareness, while GNN P is a novel implementation that can predict with no prior knowledge of the intermolecular interactions. Comprehensive evaluation demonstrated that GNN can successfully capture the binary interactions between ligand and protein's 3D structure with 0.979 test accuracy for GNN F and 0.958 for GNN P for predicting activity of a protein-ligand complex. These models are further adapted for

De novo design of protein target specific scaffold-based Inhibitors via Reinforcement Learning

“…In our approach, we utilize a scaffold-based generative model, 3D-Scaffold (Joshi et al, 2021) to produce valid 3D compounds as the actor model. For the critic model, we utilize parallel graph neural networks as a binding probability predictor (GNN P (Knutson et al, 2021)) to evaluate whether the generated compound actively binds with a target protein. This method takes the 3D structure of the protein pocket and the ligand, and predicts the probability of their interaction without any prior knowledge of the intermolecular interactions.…”