Protein fitness prediction method, device, terminal and storage medium

By constructing pairs of evolutionary constraints using pre-trained protein language models and multiple sequence alignment information, and combining them with attention mechanisms, the problem of insufficient cross-modal interaction in protein mutation fitness prediction is solved, achieving high-precision and widely applicable mutation effect analysis.

CN121393550BActive Publication Date: 2026-03-31XIDIAN UNIV
2 Cites 0 Cited by

Patent Information

Application Number
CN202511953216.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-12-23
Publication Date
2026-03-31
Estimated Expiration
2045-12-23

AI Technical Summary

Technical Problem

Existing methods lack effective cross-modal interaction mechanisms when predicting the impact of protein mutations on fitness, making it difficult to fully explore the intrinsic synergistic effects of sequence, structure, and evolutionary information, and their reliance on experimental structural data limits their application scope.

Method used

Sequence-level representations are extracted using a pre-trained protein language model, representation difference vectors are calculated, and pairwise evolutionary constraint information is constructed using multiple sequence alignment information. This is combined with attention mechanisms to model interactions and generate enhanced representations. Finally, fitness prediction results are output through a downstream prediction head, completely independent of experimental structure.

Benefits of technology

It significantly improves the accuracy and scope of protein fitness prediction, and can be widely applied to almost any protein sequence, overcoming the application bottleneck of traditional methods.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The application relates to the technical field of protein engineering and bioinformatics, and specifically provides a protein fitness prediction method and device, a terminal and a storage medium. The method comprises the following steps: extracting wild-type sequence-level representation and mutant sequence-level representation by using a pre-trained protein language model, and calculating a representation difference vector between the two; deducing coevolution coupling information from multiple sequence alignment information, and constructing pairwise evolutionary constraint information reflecting spatial proximity relationship; then modeling the interaction between the representation difference vector and the pairwise evolutionary constraint information through an attention mechanism, so that the pairwise coevolution information guides the propagation and weighting of the representation difference vector between residues to generate enhanced representation; and inputting the enhanced representation into a downstream prediction head to regress a scalar value as a fitness prediction result. The application models the interaction between the sequence-level difference vector and the pairwise evolutionary constraint, so that the non-additive effect between mutations can be accurately captured without relying on any experimental structure.
Need to check novelty before this filing date? Find Prior Art