vix.ing · top · new · best · stats · spec

Structure-Aligned Protein Language Model

2025/05/22 by Can Chen, Chen, Can, David Heurtel-Depeiges +9
Biochemistry, Genetics and Molecular Biology · #Artificial Intelligence (cs.AI) #Bioinformatics and Genomic Networks #FOS: Computer and information sciences #Genomics and Rare Diseases #Machine Learning (cs.LG) #Machine Learning in Bioinformatics

paper · pdf · doi:10.48550/arxiv.2505.16896

openalex publication_date 2025/05/22 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Protein language models (pLMs) pre-trained on vast protein sequence databases excel at various downstream tasks but often lack the structural knowledge essential for some biological applications. To address this, we introduce a method to enrich pLMs with structural knowledge by leveraging pre-trained protein graph neural networks (pGNNs). First, a latent-level contrastive learning task aligns residue representations from pLMs with those from pGNNs across multiple proteins, injecting inter-protein structural information. Additionally, a physical-level task integrates intra-protein information by training pLMs to predict structure tokens. Together, the proposed dual-task framework effectively incorporates both inter- and intra-protein structural knowledge into pLMs. Given the variability in the quality of protein structures in PDB, we further introduce a residue loss selection module that uses a small model trained on high-quality structures to select reliable yet challenging residue losses for the pLM to learn. Applying our structure alignment method as a simple, lightweight post-training step to the state-of-the-art ESM2 and AMPLIFY yields notable performance gains. These improvements are consistent across a wide range of tasks, including substantial gains in deep mutational scanning (DMS) fitness prediction and a 59% increase in P@L for ESM2 650M contact prediction on CASP16. Furthermore, we demonstrate that these performance gains are robust, scaling with model sizes from 8M to 650M and extending to different downstream tasks.

Related