Protein model quality assessment using rotation-equivariant, hierarchical neural networks

TL;DR

This paper introduces a novel deep learning method using rotation-equivariant, hierarchical neural networks to assess protein model quality, achieving state-of-the-art results without relying on physics-based or sequence alignment information.

Contribution

The authors develop a rotation-equivariant neural network that directly evaluates protein structures, advancing model quality assessment without needing external biological data.

Findings

Achieved state-of-the-art accuracy in CASP protein model scoring

Operates without physics-inspired energy terms or sequence alignments

Effective end-to-end learning from atomic-level protein structures

Abstract

Proteins are miniature machines whose function depends on their three-dimensional (3D) structure. Determining this structure computationally remains an unsolved grand challenge. A major bottleneck involves selecting the most accurate structural model among a large pool of candidates, a task addressed in model quality assessment. Here, we present a novel deep learning approach to assess the quality of a protein model. Our network builds on a point-based representation of the atomic structure and rotation-equivariant convolutions at different levels of structural resolution. These combined aspects allow the network to learn end-to-end from entire protein structures. Our method achieves state-of-the-art results in scoring protein models submitted to recent rounds of CASP, a blind prediction community experiment. Particularly striking is that our method does not use physics-inspired energy terms and does not rely on the availability of additional information (beyond the atomic structure of the individual protein model), such as sequence alignments of multiple proteins.