ABSTRACT
Multiple chromosomal aberrations, nucleotide substitutions, and epigenetic modifications may occur in human cancer cells, which drive malignant transformation. The Cancer Genome Atlas (TCGA) project aims to promote large-scale multi-dimensional analysis of these molecular characteristics in human cancer and rapidly provide data to researchers. In this study, we introduce four flow paths of the production of TCGA data, the collections of various cancer types, the data category and level, and the standardized pipeline of data analysis, as well as several existing data analytical tools. We used ovarian cancer as an example to introduce the application of the TCGA data in the analyses of mutation, copy number, analysis, and expression. We summarized the important findings of glioblasto-ma by TCGA teams.
ABSTRACT
<p><b>OBJECTIVE</b>To explore the pathogenesis of ovarian cancer from the perspective of molecular genetic variation and changes in mRNA expression profiles.</p><p><b>METHOD</b>The data of DNA copy number and mRNA expression profiles of high-grade serious ovarian cancer were obtained from TCGA. The significant copy number variation regions were identified using the bioinformatics tool GISTIC, and the differentially expressed genes in these regions were identified using the samr package of SAM. The selected genes were subjected to bioinformatics analysis using GSEA tools.</p><p><b>RESULTS</b>GISTIC analysis identified 45 significant copy number amplification regions in ovarian cancer, and SAM and Fisher's exact test found that 40 of these genes showed altered expression levels. GSEA enrichment analysis revealed that most of these genes were reported in several published studies describing genetic study of tumorigenesis.</p><p><b>CONCLUSION</b>An integrative bioinformatics study of DNA copy number variation data and microarray data can identify genes involved in tumor pathogenesis. and offer new clues for studying early diagnosis and therapeutic target of ovarian cancer.</p>