Study Of The Relevance Between Gene Expression And Gene Body Methylation In Primordial Germ Cells Of Mouse

DNA methylation is one of the most common types of epigenetic modification.Methylation in the regulating area upstream of genes can regulate expression ofdownstream genes. Generally DNA methylation in transcription initiation region isrelated with gene silencing, which is mainly because that methylation can change thethree-dimensional conformation of DNA, thus affecting the binding of transcriptionfactor; or methylated DNA can interact with the methylation binding proteins toinhibit the initiation of transcription. In recent years, more and more studies havefound, the gene body regions located downstream of the transcription start site alsohave different levels of methylation, and the methylation and gene expression waspositively correlated. To explain this paradox, it is expected that the methylation onlyaffects the transcription initiation, but will not affect the elongation of transcription.So what is the relationship between body methylation and gene expression? Howmany genes’ expression level was affected by body methylation? If there is somethingin common within genes whose expression level are highly correlated with genesbody methylation? In this study, to study these issues, we use genome-widemethylation sequencing and transcriptome sequencing data, to conduct acomprehensive analysis of body methylation and gene expression in mouse primordial germ cells of different developmental periods. We found that the methylation of maleand female cells are very difference during expression changes in the process, at thesame time, the relationship between the level of methylation in coding region andexpression is more significant; on the other hand, we found that there are about300genes whose methylation are highly correlated with expression level, and the contentof CG of these genes is generally lower than that of other genes; finally, we found thatthe relationship between methylation and expression of genes with single transcript ismore close, and genes containing CpG island in their bodies have lower expressionviariation than genes without CpG island in their bodies. These results support theconclusion that the body methylation and expression was positive related. Meanwhile,it provides some new ideas for further study of their relation, and provides a basis forvalidation in more species or sample. Hepatocellular carcinoma (HCC) is the most common type of liver cancerworldwide and one of the deadliest cancers in Asia. Given the close associationbetween Hepatitis B virus and Hepatitis B, it can also be an important factor in HCC.There are90%HCC patients who have ever been infected by HBV. But at present,effective targets for HCC therapy in clinic are still limited, and we still do not knowhow many genes there are playing key roles in tumorgenesis and progression. The‘‘guilt by association’’ rule suggests that interacting proteins share the same or similarfunctions and hence may be involved in the same pathway. This assumption can beused to identify disease related genes from protein association networks constructedfrom existing PPI data. Here we develop a computational method to identifyhepatocellular carcinoma related genes based on k-th shortest paths in the protein-protein interaction (PPI) network (we set k1,2in this study). Finally, we found33genes whose p-values were less than0.05, and most of them have been reported to beinvolved in HCC tumorigenesis and development. We also predict some new geneswhich have never been reported to be involved in HCC. The results can provide a newreference for research into HCC oncogenesis and for development of new strategiesfor HCC clinical therapies. The method can also be used to predict other diseasegenes.