Mining Overlapping Protein Complexes in PPI Network Based on Granular Computation in Quotient Space

Abstract

Proteins complexes play a critical role in many biological processes. The existing protein complex detection algorithms are mostly cannot reflect the overlapping protein complexes. In this paper, a novel algorithm is proposed to detect overlapping protein complexes based on granular computation in quotient space. Firstly, problems are expressed by quotient space and different quotient space embodies the quotient set of different granular. Then the method estimates the relationship between particles to make up for the inadequacy of data in combination with the PPI data and Gene Ontology data, deals with the network based on quotient space theory. Graining the network to construct the quotient space and merging the particles layer by layer. The final protein complexes is obtained after purification. The experimental results on Saccharomyces cerevisiae and Homo sapiens turned out that the proposed method could exploit protein complexes more accurately and efficiently.