Recognition method of protein complexes

Abstract

The invention discloses a recognition method of protein complexes. The method comprises following steps: constructing a weighted protein-protein interaction network, wherein weight value represents similarity between gene representing modes corresponding to proteins with codes interacting each other; recognizing nucleuses of protein complexes based on the weighted protein-protein interaction network; recognizing accessory proteins for the nucleuses of protein complexes based on the number of nuclear proteins interacting with non-nuclear proteins and weight; and combining recognized nucleuses and recognized accessory proteins to recognize the protein complexes. The method does not take internal structures of protein complexes into consideration and utilizes the feature of code interaction with high protein gene expression. Therefore, the method is capable of recognizing protein complexes with overlapping structures and the recognized protein complexes reflect actual internal structures of protein complexes.

Description

technical field [0001] The invention relates to a protein network, and more specifically relates to a method for identifying protein complexes. Background technique [0002] Large-scale protein network is a special complex network with obvious modular organization structure, and these modular structures often correspond to protein complexes or functional modules. Mining protein complexes from protein networks is of great help in studying protein functions and explaining specific biological processes. Mining protein complexes in protein networks is a challenging topic, and until now there is no clear and accepted definition of protein complexes. The study found that protein complexes often appear as some dense subgraphs in the network. Therefore, most researchers transform the mining of protein complexes into the problem of identifying dense subgraphs in protein networks, and a large number of protein complex mining algorithms based on protein networks have been produced. ...

AI Technical Summary

Problems solved by technology

Although these methods have contributed to improving the accuracy of protein complex mining algorithms, these methods either use gene expression data to simply denoise protein interaction data, or weight protein interactions to construct a weighted network, and do not The core-appendage structure of protein complexes is effectively combined with gene expression data, especially gene co-expression data is not effectively used for mining protein complexes

Images