Protriever: End-to-End Differentiable Protein Homology Search for Fitness Prediction

TL;DR

Protriever is a novel end-to-end differentiable framework for protein homolog retrieval that improves fitness prediction accuracy and efficiency over traditional MSA-based methods, enabling scalable and adaptable protein modeling.

Contribution

It introduces Protriever, a flexible, task-agnostic, end-to-end differentiable approach that replaces costly MSA-based retrieval with efficient vector search for protein homologs.

Findings

Achieves state-of-the-art protein fitness prediction performance.

Operates two orders of magnitude faster than MSA-based methods.

Flexible and adaptable to different retrieval strategies and databases.

Abstract

Retrieving homologous protein sequences is essential for a broad range of protein modeling tasks such as fitness prediction, protein design, structure modeling, and protein-protein interactions. Traditional workflows have relied on a two-step process: first retrieving homologs via Multiple Sequence Alignments (MSA), then training models on one or more of these alignments. However, MSA-based retrieval is computationally expensive, struggles with highly divergent sequences or complex insertions & deletions patterns, and operates independently of the downstream modeling objective. We introduce Protriever, an end-to-end differentiable framework that learns to retrieve relevant homologs while simultaneously training for the target task. When applied to protein fitness prediction, Protriever achieves state-of-the-art performance compared to sequence-based models that rely on MSA-based homolog retrieval, while being two orders of magnitude faster through efficient vector search. Protriever is both architecture- and task-agnostic, and can flexibly adapt to different retrieval strategies and protein databases at inference time -- offering a scalable alternative to alignment-centric approaches.