Deep Analysis Methods And Application Of MiRNA Expression Profiles

With the development of biology highthroughput technology,more and more coding and non-coding gene expression data become avaliable,how to interpret the biological meaning behind these data become the hot topic of life science field.micro RNAs(mi RNAs)are a class of small non-coding RNAs that with 18~25nt in length,and widely exist in eukaryotic cells.Although mi RNA can not code protein,they can prevent the targets' translation or promote m RNA degradation at the post-transcriptional level,thereby regulatting for typical biological function.Previous studies show that differentially expressed mi RNAs play very important roles during the formation and development of multiple complex diseases,this brings the hope that mi RNA can be disease biomarker and therapeutic target.With the available of mi RNA expression profile,this let us can systematically study the function of mi RNA in complex disease.Currently,the most methods for analizing mi RNA expression profiles,the first step of which is to find the differentially expressed mi RNAs,and then predict the targets of these mi RNAs,further obtain the biological function that those mi RNAs regulated.However,these methods only think of the differentially expression of single mi RNA,and ignore that the multiple to multiple interactions between mi RNA and m RNA.This study is mainly based on the mi RNA expression profiles,or integrating mi RNA and m RNA expression profiles,and develop a practical method to compute the repression score of the whole mi RNA expression profile on each m RNA.1 System analysis method of mi RNA expression profiles and application.we propose a method for systematically analyzing the mi RNA expression profile.The binding ability between different mi RNA-m RNA pairs are with difference,this coefficient in some way represents the regulatory coefficient between mi RNA and its target m RNA.By integrating the mi RNAs expression and the regulatory coefficient,we propose a method for measuring the impact of mi RNA to its target m RNA.Based on this method,firstly,we can know which m RNAs are differentially regulated by the whole mi RNA expression profile,secondly,for specific m RNA,we can know that which mi RNAs have significant differential impact on it.Thus,this method can be used for screening of mi RNA biomarker and therapeutic drug target.In this study,we apply this method for comprehensively analyze mi RNA expression profile.2 Deep analysis methods and application of mi RNA expression based on m RNA experimental data.With the increasingly mature of biological highthroughput technology,along with the lower of sequencing cost,this make it possible for integrated analysis based on multiple omics data.To be more accuratly analyze the impact of mi RNA on its target,we propose a project for comprehensively analyzing the mi RNA expression profiles based on m RNA experimental data.In the proposed math model,during computting the repression score of mi RNA on its targets,we not only take into account of mi RNA expression profile but also the expression of its targets m RNAs.Based on the mi RNA and m RNA expression profiles of colorectal cancer,we apply the new method on these samples,results show that the m RNAs that significantly regulated by the whole mi RNA expression profiles are significantly enriched in multiple cancer related pathways including MAPK.Furthermore,to study the important role of mi RNAs in the systematic out-of-balance diseases,we summrize the genes that correlated with cell fates,and than study the impact of mi RNAs on the cell fate genes from network level.Through analyzing the expression profiles of colorectal cancer,non-small cell lung cancer and liver cancer,results show that differentially changed mi RNAs mainly promote the cell proliferation,while inhibit the cell apoptosis.This provides a new veiw for study the important of mi RNA in the development of cancer.3 Method of accurate screening and analysis of mi RNA based on m RNA local dynamic network.Based on the mi RNAs high throughput sequencing data of both colorectal cancer stem cells and normal cultured colorectal cancer cells,we first filter the m RNAs that differentially repressed by the mi RNAs,KEGG pathway analysis shows that these m RNAs are enriched in multiple cancer relevant signalling pathways including Signaling pathways regulating pluripotency of stem cells.Then,based on the Signaling pathways regulating pluripotency of stem cells,we constructed a mi RNAs/m RNAs network centering on Oct-4,Nanog and Sox2,which are considered as to be correlated to the self-renew of cancer stem cells.Further,based on the mi RNAs/m RNAs network,the impact index of mi RNAs that influence the cancer stem self-renew were computed.For further verification of mi RNAs that influence the self-renew,a generalized logic model was adopted,results show that those mi RNAs that we screened out indeed have a influence to the stem cells self-renew.In the end,through cell cloning and cell cycle biological experiments,we demonstrated that the over-expression of mi R-92a-3p inhibit cell cloning and cell cycle.In summary,the method proposed based on mi RNA or integratting mi RNA and m RNA expression profile,is helpful for study the mi RNA characteristic of cancer or other complex disease.First,we can comprehensively evaluate the signalling pathways or GO modules that regulated by mi RNAs.In the next place,from the enriched pathways,we can find out that which mi RNAs are related to specific pathway,which can provide important clues for screening mi RNA biomarkers or drugs.In the end,based on the mi RNAs/m RNAs regulatory network,we can systematically study the impact on target genes by mi RNA perturbation,which provides theoretical prediction for mi RNA drug screening.