A PLMs based protein retrieval framework

TL;DR

This paper introduces a novel protein retrieval framework leveraging protein language models and vector indexing to improve retrieval of both similar and dissimilar proteins, surpassing traditional sequence similarity methods.

Contribution

The proposed framework mitigates sequence similarity bias by embedding proteins in a high-dimensional space using PLMs and employing an accelerated vector database for efficient retrieval.

Findings

Effectively retrieves both similar and dissimilar proteins.

Identifies proteins overlooked by traditional methods.

Enhances protein mining and biological research.

Abstract

Protein retrieval, which targets the deconstruction of the relationship between sequences, structures and functions, empowers the advancing of biology. Basic Local Alignment Search Tool (BLAST), a sequence-similarity-based algorithm, has proved the efficiency of this field. Despite the existing tools for protein retrieval, they prioritize sequence similarity and probably overlook proteins that are dissimilar but share homology or functionality. In order to tackle this problem, we propose a novel protein retrieval framework that mitigates the bias towards sequence similarity. Our framework initiatively harnesses protein language models (PLMs) to embed protein sequences within a high-dimensional feature space, thereby enhancing the representation capacity for subsequent analysis. Subsequently, an accelerated indexed vector database is constructed to facilitate expedited access and retrieval of dense vectors. Extensive experiments demonstrate that our framework can equally retrieve both similar and dissimilar proteins. Moreover, this approach enables the identification of proteins that conventional methods fail to uncover. This framework will effectively assist in protein mining and empower the development of biology.