Protein function prediction method based on network propagation

Abstract

The invention discloses a protein function prediction method based on network propagation, and the method comprises the following steps: building an undirected weighted network according to a proteininteraction network topological structure, and enabling the undirected weighted network to be represented as an undirected graph adjacency matrix M; performing calculating to obtain an initial propagation distance matrix MI according to the protein structural domain information and the protein compound information; iteratively calculating a network propagation distance between proteins according to the undirected graph adjacency matrix M and the initial propagation distance matrix MI to obtain a network propagation matrix MP; gradually adding adjacent nodes of the test protein according to thenetwork propagation matrix MP to form function modules, and forming a candidate function list by a plurality of function modules; scoring and sorting the candidate functions in the candidate functionlist, and selecting a plurality of function annotation test proteins with the top rank. According to the method, the fusion mode of multi-source biological data in protein function prediction methodresearch is improved, the negative influence of false positive and false negative in the interaction network on prediction is reduced, and the accuracy of protein function prediction is greatly improved.

Description

technical field [0001] The invention relates to the technical field of protein function prediction, in particular to a protein function prediction method based on network propagation. Background technique [0002] Accurate annotation of protein function is the key to understanding life at the molecular level, which has a huge impact on biomedicine and pharmaceuticals. As genomes of many species have been sequenced, discrepancies between sequence data and their functional annotations have increased. The sequence data with functional annotations in the UniProt database is less than 14%, and the sequence data is still growing exponentially. Early predictions of protein function were based on experimental methods such as gene knockout, targeted mutation and inhibition of gene expression. These experimental methods require a lot of experimental and labor costs, and are not suitable for annotating important parts of proteins, which makes the gap between sequence and function inc...

AI Technical Summary

Problems solved by technology

[0013] The present invention provides a protein function prediction method based on network propagation to solve the problem of poor protein function prediction performance in the prior art

Images