Cluster-based Gene Differential Coexpression Analysis

As the biological science and technology rapid development,bioinformatics has emerged.It is the study of biology and biological information content in the system and information flow to the integrated system discipline.With the emergence of gene chip technology,attracted many researchers to join in this area.The DNA microarray data to differential coexpression analysis,find genes associated with dis-ease is the most popular research projects of the current bioinformatics,it is also a new method for understanding and treating diseases.Now a lot of genetic diseases of medicine pathology also is not very understand-ing,so differential coexpression,analysis gradually be applied to search for genes associated with disease center.This paper first to analyze differentially expressed for gene expression dataset,employing R package to screening 356 down-regulated differentially expressed genes and 36 up-regulated differentially expressed genes.By comparing under normal and disease states of two differentially expressed gene clus-tering figure,we will find some differentially expressed genes of class change,then we enrichment analysis of these genes of KEGG Pathway,down-regulated DEGs are mainly involved in the the pathways related to“metabolism”.However,the up-regulated DEGs are included in Pentose phosphate pathway,insulin signaling pathway and so on,besides two metabolism-associated pathways.Next we differen-tial coexpression analysis for differentially expressed genes,current Differential coex-pression analysis(DCEA)methods mostly adopt a gene connectivity-based strategy to estimate differential coexpression,although it simplifies the calculation,and fails to differentiate significant differential coexpression changes from those trivial ones.The correlation-reversal is easily missed although it probably indicates remarkable biological significance.We developed two link-based quantitative methods to identi-fy differentially coexpressed Genes,the method compared with classic methods,the effectiveness of this method.Finally will be difference coexpressed genes annotation information output.