Prediction Study On Tumor Markers For Prostate Cancer And Colon And Rectal Cancer Based On Bioinformatics

According to the latest national cancer statistics released by the National Cancer Center in January 2019,the number of malignant tumors in China in 2015 was approximately 3.929 million,and the number of deaths was approximately 2.338 million.This means that on average more than 10,000 people are diagnosed with cancer every day,and 7.5 people are diagnosed with cancer every minute.This data shows a continuous upward trend compared with historical data.For more than 10 years,the incidence of malignant tumors has increased by about 3.9% each year,and the mortality rate has increased by 2.5% each year.Among them,colorectal cancer has become one of the malignant tumors that endanger the health of our population,and the remaining are: lung cancer,liver cancer,upper digestive system tumors and female breast cancer.In addition,among men,prostate cancer has been on the rise in recent years,ranking sixth among men.With the widespread application of gene chips and high-throughput sequencing technology,a large amount of gene expression profile data and sequencing data have been generated.The emergence of a new generation of artificial intelligence technology can provide algorithms and technical foundations for analyzing these biological data,thereby achieving prediction of tumor-related markers and drug targets.The verification for these results can be verified through wet experiments.But in the era of big data,these results can be verified through multiple public bioinformatics databases.This comprehensive analysis is a step forward in revealing the biological processes of cancer.This study will use bioinformatics methods and tools,and integrate gene expression profile data,methylation data and related omics data in the GEO and TCGA databases,and write code in R.The differentially expressed genes were screened to construct an interaction network of these key genes,and finally the key genes and tumor suppressor genes that affect prostate and colorectal cancer were screened.The relevant data of tumor patients in this study came from gene expression profiles(GEPs)in the GEO database,Level 3 data in the Broad GDCA Firehose database,and clinical data in the TCGA database.The development platform for this research is RStudio 1.453 with Affy,methylumiIlluminaHumanMethylation-450 kmanifest,limma,minfi,watermelon,IlluminaHumanMethylation450 kanno.ilmn12.hg19,WGCNA,dynamicTreeCut and fastcluster package.Then,PPI analysis was performed on the differential expression results,and a differential gene interaction network was constructed,and the key genes were screened in conjunction with graph theory related algorithms.Furthermore,functional enrichment analysis of differential genes was performed by DAVID online tools,and important signal pathways were obtained using the KEGG database.Combined with immunoinfiltration and proteomics database,the results of these studies were analyzed by multi-omics.Finally,the key genes,tumor suppressor genes,and drug targets were verified by integrating multiple publicly available biological information databases.In the first section,we conducted an intersection analysis of the DNA methylation profiling and gene expression profiling of prostate cancer to screen out key genes and candidate tumor suppressor genes.Candidate tumor suppressor genes are IKZF1,PPM1 A,FBP1,SMCHD1,ALPL,CASP5,PYHIN1,DAPK1,and CASP8.The key genes are FGFR1,FGF13 and CCND1.In the second section,we used machine learning algorithms to analyze the gene expression profile of prostate cancer,and obtained key genes and drug targets.These tumor markers are helpful for the study of their molecular mechanisms.The key genes are UBE2 C,CCNB1,TOP2 A,TPX2,CENPM,KIAA0101,F5,APOE,NPY and TRIM36.In the third part of the article,we use the dynamic network biomarker(DNB)algorithm of the tumor to obtain the key gene MYC of four molecules of colorectal cancer tumors.The results show that MYC can be used as a dynamic marker for the diagnosis and treatment of colorectal cancer.The tumor suppressor genes are ZBTB16,MAL,LIFR and SLIT2.