A Point Cloud-Based Deep Learning Strategy for Protein-Ligand Binding Affinity Prediction

TL;DR

This study introduces a novel approach using point cloud-based deep learning models, PointNet and PointTransformer, for predicting protein-ligand binding affinity, achieving competitive accuracy and interpretability.

Contribution

First application of point cloud deep learning models to protein-ligand affinity prediction, demonstrating their effectiveness and interpretability in this domain.

Findings

PointNet and PointTransformer achieved Pearson correlations of 0.831 and 0.859.

Models can visualize key interaction atoms in protein-ligand complexes.

PointTransformer features can enhance machine learning models like XGBoost.

Abstract

There is great interest to develop artificial intelligence-based protein-ligand affinity models due to their immense applications in drug discovery. In this paper, PointNet and PointTransformer, two pointwise multi-layer perceptrons have been applied for protein-ligand affinity prediction for the first time. Three-dimensional point clouds could be rapidly generated from the data sets in PDBbind-2016, which contain 3 772 and 11 327 individual point clouds derived from the refined or/and general sets, respectively. These point clouds were used to train PointNet or PointTransformer, resulting in protein-ligand affinity prediction models with Pearson correlation coefficients R = 0.831 or 0.859 from the larger point clouds respectively, based on the CASF-2016 benchmark test. The analysis of the parameters suggests that the two deep learning models were capable to learn many interactions between proteins and their ligands, and these key atoms for the interaction could be visualized in point clouds. The protein-ligand interaction features learned by PointTransformer could be further adapted for the XGBoost-based machine learning algorithm, resulting in prediction models with an average Rp of 0.831, which is on par with the state-of-the-art machine learning models based on PDBbind database. These results suggest that point clouds derived from the PDBbind datasets are useful to evaluate the performance of 3D point clouds-centered deep learning algorithms, which could learn critical protein-ligand interactions from natural evolution or medicinal chemistry and have wide applications in studying protein-ligand interactions.