E(3) equivariant graph neural networks for robust and accurate protein-protein interaction site prediction

Artificial intelligence-powered protein structure prediction methods have led to a paradigm-shift in computational structural biology, yet contemporary approaches for predicting the interfacial residues (i.e., sites) of protein-protein interaction (PPI) still rely on experimental structures. Recent studies have demonstrated benefits of employing graph convolution for PPI site prediction, but ignore symmetries naturally occurring in 3-dimensional space and act only on experimental coordinates. Here we present EquiPPIS, an E(3) equivariant graph neural network approach for PPI site prediction. EquiPPIS employs symmetry-aware graph convolutions that transform equivariantly with translation, rotation, and reflection in 3D space, providing richer representations for molecular data compared to invariant convolutions. EquiPPIS substantially outperforms state-of-the-art approaches based on the same experimental input, and exhibits remarkable robustness by attaining better accuracy with predicted structural models from AlphaFold2 than what existing methods can achieve even with experimental structures. Freely available at https://github.com/Bhattacharya-Lab/EquiPPIS, EquiPPIS enables accurate PPI site prediction at scale.Author summary: Predicting how proteins interact and characterizing the interacting residues at the protein-protein interaction interface (i.e., sites) is of central importance to understanding various biological processes actuated by protein-protein interactions (PPI). Despite the remarkable recent progress in protein structure prediction driven by artificial intelligence, existing approaches for PPI site prediction still rely on experimental input. This paper presents an E(3) equivariant graph neural network approach for PPI site prediction that takes into account symmetries naturally occurring in 3-dimensional space and transforms equivariantly with translation, rotation, and reflection. Rigorous experimental validation shows that our method attains substantially improved accuracy and robustness over the existing approaches. Moving beyond what is currently possible with only experimental input, our method enables large-scale PPI site prediction using predicted structural models from AlphaFold2 without compromising on accuracy. An open-source software implementation of EquiPPIS, licensed under the GNU General Public License v3, is freely available at https://github.com/Bhattacharya-Lab/EquiPPIS.