Beyond ESM2: Graph-Enhanced Protein Sequence Modeling with Efficient Clustering

TL;DR

This paper introduces a novel protein modeling approach that enhances the ESM2 framework by integrating protein family classification and clustering techniques, resulting in improved global and local protein representations for biological analysis.

Contribution

It combines community clustering with a contextual prediction task to significantly improve protein representations beyond existing models like ESM2.

Findings

Achieved state-of-the-art results in downstream protein tasks

Enhanced global protein representations through clustering

Improved amino acid prediction accuracy

Abstract

Proteins are essential to life's processes, underpinning evolution and diversity. Advances in sequencing technology have revealed millions of proteins, underscoring the need for sophisticated pre-trained protein models for biological analysis and AI development. Facebook's ESM2, the most advanced protein language model to date, leverages a masked prediction task for unsupervised learning, crafting amino acid representations with notable biochemical accuracy. Yet, it lacks in delivering functional protein insights, signaling an opportunity for enhancing representation quality.Our study addresses this gap by incorporating protein family classification into ESM2's training.This approach, augmented with Community Propagation-Based Clustering Algorithm, improves global protein representations, while a contextual prediction task fine-tunes local amino acid accuracy. Significantly, our model achieved state-of-the-art results in several downstream experiments, demonstrating the power of combining global and local methodologies to substantially boost protein representation quality.