Extracting Inter-Protein Interactions Via Multitasking Graph Structure Learning

TL;DR

This paper introduces MgslaPPI, a novel multitask graph attention-based method for predicting protein-protein interactions that incorporates structural information and auxiliary tasks to improve accuracy.

Contribution

The paper presents a new PPI prediction framework that combines graph attention, multitask learning, and protein structural information, outperforming existing methods.

Findings

MgslaPPI significantly outperforms state-of-the-art methods.

The use of auxiliary tasks enhances prediction accuracy.

Graph attention effectively captures protein structural features.

Abstract

Identifying protein-protein interactions (PPI) is crucial for gaining in-depth insights into numerous biological processes within cells and holds significant guiding value in areas such as drug development and disease treatment. Currently, most PPI prediction methods focus primarily on the study of protein sequences, neglecting the critical role of the internal structure of proteins. This paper proposes a novel PPI prediction method named MgslaPPI, which utilizes graph attention to mine protein structural information and enhances the expressive power of the protein encoder through multitask learning strategy. Specifically, we decompose the end-to-end PPI prediction process into two stages: amino acid residue reconstruction (A2RR) and protein interaction prediction (PIP). In the A2RR stage, we employ a graph attention-based residue reconstruction method to explore the internal relationships and features of proteins. In the PIP stage, in addition to the basic interaction prediction task, we introduce two auxiliary tasks, i.e., protein feature reconstruction (PFR) and masked interaction prediction (MIP). The PFR task aims to reconstruct the representation of proteins in the PIP stage, while the MIP task uses partially masked protein features for PPI prediction, with both working in concert to prompt MgslaPPI to capture more useful information. Experimental results demonstrate that MgslaPPI significantly outperforms existing state-of-the-art methods under various data partitioning schemes.