Research On Core Genes Involved In Tumor Metastasis Based On Complex Network

Tumor metastasis is a complicated and multistage process and causes most cancer-related deaths.The study of related genes in the process of metastasis has become a focused area in the molecular biology of tumor metastasis.The identification of candidate genes involved in tumor metastasis can be helpful to the discovery of anti-drug molecular targets and the development of personalized treatment,and it is urgent and significant for the diagnosis and treatment of cancer cases.Compared with the expensive and time-consuming biological experimental methods,it is more effective to find tumor metastasis-related genes through computational models.In this study,the construction of Protein-Protein Interaction Networks(PPIN),combined with complex network methods,carried out the research on the core genes of tumor metastasis and the mechanism of metastasis from the molecular level.The specific research contents are as follows:(1)In response to the discovery of core genes involved in tumor metastasis,we proposed a method based on random walk with restart(RWR)algorithm and applied it to identify genes related to lung cancer bone metastasis.First,the verified lung cancer and bone cancer genes were used as seed nodes,and the RWR algorithm was applied on PPIN to preselect the gene.Then we performed permutation test to eliminate the interference of network structure.In addition,these genes were further screened by interaction test and enrichment test.Finally,12 potential core genes for lung cancer bone metastasis were obtained.The results of gene function analysis proved that 10 of the inferred genes were contribute to the process of lung cancer bone metastasis,thereby suggesting the effectiveness of our method,as well as revealing the potential molecular processes that these core genes may participate in.(2)For the further studies on the mechanism of tumor metastasis,a method combining RWR algorithm and betweenness centrality was proposed to retrieve genes involved in breast cancer bone metastasis.Firstly,the RWR algorithm and permutation test were used to preselect genes and obtain candidate genes.Then we constructed a key subnet of candidate genes and ranked the nodes by betweenness centrality value.Finally,we implemented the interaction test and enrichment test for further screening to ensure the accuracy of the results,30 core genes for breast cancer bone metastasis were obtained.Further analysis based on interaction analysis,enrichment analysis,subsistence analysis and gene function indicated that 26 core genes contribute to the initiation or progression of breast cancer bone metastasis,the other 4 genes might be potential breast cancer bone metastasis-related genes,which deserves further investigation.The results of this study revealed the process of tumor metastasis from the molecular level.It is hoped that this contribution would provide target genes for clinical trials of tumor metastasis.In addition,the effectiveness of the proposed model was verified through the analysis of core genes,which provides new ideas for the study of pathogenic genes involved in complex diseases.