Towards Multiscale Graph-based Protein Learning with Geometric Secondary Structural Motifs

TL;DR

This paper introduces a multiscale graph-based framework for protein learning that captures local and global structural features efficiently, improving accuracy and reducing computational costs.

Contribution

It proposes a hierarchical graph representation with secondary structure motifs and a dual GNN approach, enhancing multiscale modeling of proteins.

Findings

Improves prediction accuracy on protein structure benchmarks.

Reduces computational cost compared to existing methods.

Preserves structural information through hierarchical design.

Abstract

Graph neural networks (GNNs) have emerged as powerful tools for learning protein structures by capturing spatial relationships at the residue level. However, existing GNN-based methods often face challenges in learning multiscale representations and modeling long-range dependencies efficiently. In this work, we propose an efficient multiscale graph-based learning framework tailored to proteins. Our proposed framework contains two crucial components: (1) It constructs a hierarchical graph representation comprising a collection of fine-grained subgraphs, each corresponding to a secondary structure motif (e.g., $\alpha$-helices, $\beta$-strands, loops), and a single coarse-grained graph that connects these motifs based on their spatial arrangement and relative orientation. (2) It employs two GNNs for feature learning: the first operates within individual secondary motifs to capture local interactions, and the second models higher-level structural relationships across motifs. Our modular framework allows a flexible choice of GNN in each stage. Theoretically, we show that our hierarchical framework preserves the desired maximal expressiveness, ensuring no loss of critical structural information. Empirically, we demonstrate that integrating baseline GNNs into our multiscale framework remarkably improves prediction accuracy and reduces computational cost across various benchmarks.