Growing ecosystem of deep learning methods for modeling protein$\unicode{x2013}$protein interactions

TL;DR

This paper reviews the expanding use of deep learning techniques in modeling protein–protein interactions, emphasizing diverse methods, recent successes, challenges, and future directions in understanding and engineering these biological processes.

Contribution

It provides a comprehensive overview of biophysically-informed deep learning models for protein interactions, highlighting their applications, advantages, and limitations.

Findings

Representation learning captures complex interaction features.

Geometric deep learning predicts protein structures and complexes.

Generative models enable de novo protein design.

Abstract

Numerous cellular functions rely on protein$\unicode{x2013}$protein interactions. Efforts to comprehensively characterize them remain challenged however by the diversity of molecular recognition mechanisms employed within the proteome. Deep learning has emerged as a promising approach for tackling this problem by exploiting both experimental data and basic biophysical knowledge about protein interactions. Here, we review the growing ecosystem of deep learning methods for modeling protein interactions, highlighting the diversity of these biophysically-informed models and their respective trade-offs. We discuss recent successes in using representation learning to capture complex features pertinent to predicting protein interactions and interaction sites, geometric deep learning to reason over protein structures and predict complex structures, and generative modeling to design de novo protein assemblies. We also outline some of the outstanding challenges and promising new directions. Opportunities abound to discover novel interactions, elucidate their physical mechanisms, and engineer binders to modulate their functions using deep learning and, ultimately, unravel how protein interactions orchestrate complex cellular behaviors.