Differential Analysis Of Gene Expression Based On Gene Co-expression Networks

Network modeling has been proved to be a basic tool for analysis of working principle of the cells.People's understanding of biological processes has been completely change,and in the discovery of disease biomarkers has made great progress.Researchers have been committed to using functional genomics data sets produced by high throughput technology to rebuild the various types of biochemical networks.Gene co-expression network as a kind of gene regulatory network,it is made up of genes as a node,the interaction between genetic relationship as the edge.It can be used to search for hub to discover new genes and gene module of key genes that cause cancer gene subtype,cancer,etc.,and provides help for medical research.Many biological research fields such as drug design need gene regulatory networks,to provide a clear understanding of cellular processes in living cells and understanding.This is because of the interaction between genes and their products play an important role in the process of many molecules.Gene regulatory networks can be used as the researchers to observe the relationship between genetic blueprint.Because of its importance,several calculation methods have been proposed to infer gene regulatory networks in gene expression data.DNA,RNA,proteins,and metabolites between complex regulating network how to realize the phenotype and function of organisms.Genome sequencing calculation method allows us to read the section of data analysis and molecular how to generate a comprehensive framework produces morphological phenotypes.Gene co-expression network,together with difference correlation analysis,provide the most generic genetic interaction to explore tools,using gene expression patterns,to determine the potential association and modular genetic differences between interaction and network.Combination of different types of biological data network can help to cut the whole of the complexity of biological information for discrete and can explain the data fragments,they are assuming that the starting point for building and testing.The identification of condition specific sub-networks from gene expression profileshas important biological applications,ranging from the selection of disease-related biomarkers to the discovery of pathway alterations across different phenotypes.Although many methods exist for extracting these sub-networks,very few existing approaches simultaneously consider both the differential expression of individual genes and the differential correlation of gene pairs,losing potentially valuable information in the data.In this article,we propose a new method,which employs a scoring function that jointly measures the condition-specific changes of both ‘nodes'(individual genes)and‘edges'(gene–gene co-expression).It uses the genetic algorithm to search for the single optimal sub-network which maximizes the scoring function.We applied this method to both simulated datasets with various differential expression patterns,and three real datasets,one prostate cancer dataset,a second one from the across-tissue comparison of morbidly obese patients and the other from the across-population comparison of the HapMap samples.Compared with previous methods,this method is more powerful in identifying truly significant sub-networks of appropriate size and meaningful biological relevance.