Cancer Related Signaling Pathway Analysis

With the deterioration of the ecological environment or other problems,the incidence of cancer continues to rise worldwidely,especially in China.Half of the cancer patients are in Asia,while most of these cancer patients in Asia are concentrated in China.A large number of studies have indicated that the occurrence of most cancers is related to the dysfunction of the signaling pathways.Analysis cancer related signaling pathways can help researchers studing targeted drugs for cancer treatments.In this paper,we mainly identify cancer related signaling pathways in two ways.Firstly,original signaling pathway impact analysis method(SPIA)simply attributed the relationship between genes to activation(+1)and inhibitory(-1),and in the strength of relationship between genes in actual cell network is not taken into account.In view of this problem,we proposed two new methods: signaling pathway impact analysis method based on Pearson correlation coefficient or mutual information.Results in the colon cancer dataset,lung cancer dataset and pancreatic cancer datasets show that the method based on the Pearson correlation coefficient or mutual information can identify more potential cancer-related signaling pathways than the original signaling pathway impact analysis method,and the method based on mutual information perform better than that based on the Pearson correlation coefficient.Secondly,in recent years,researchers believe that the occurrence and development of cancer is not always due to the malfunction of whole signaling pathway,but more likely due to the change of fuctionalities of a local area in signaling pathways.So the subpathway analysis method causes the attention of researchers.The definition of subpathways of the existing subpathway analysis methods is not uniform.Some define subpathway as a groups or clusters of differentially expressed genes,and some define a linear signaling subpath as a subpathway.In KEGG,most signaling pathways link with some specific biological fuctionalities at the end of pathway,so this paper defined the subnetwork including all the nodes which point to a specific fuctionalities as a subpathway.Applying this method in the colon cancer dataset,lung cancer dataset and pancreatic cancer dataset,results show that this method significantly improves the identification accuracy of cancer-related signaling pathways.And we found in these fuctionalities,some are significant differences in common in the three kinds of cancer,and some is related with specific cancer.This study provides an important guidance for the further understanding of the relationship between the fuctionalities in biological pathways and cancer.