Reveal the Correlations between Genomic and Transcriptomic Alterations Using Public Cancer Database

Large-scale cancer sequencing projects such as The Cancer Genome Atlas (TCGA), Therapeutically Applicable Research to Generate Effective Treatments (TARGET) or International Cancer Genome Consortium (ICGC) have generated sequencing data from thousands of cancer samples for more than 30 cancer types. Using the DNA alterations, mRNA expression profile, miRNA profile, clinical data and epigenetic modifications from the same individual, one can integrate all these different omics together for the very first time. Previously, cancer genomic studies demonstrated that genomic alterations including somatic mutations and genomic rearrangement are responsible for alterations in downstream transcriptomes and proteomes and ultimately carcinogenesis, cancer development. For some tumor suppressors or oncogenes, significant positive correlations were found between gene dosage and gene expression levels and some somatic mutations were correlated with transcriptomic dysregulation or changes in alternative splicing isoform preference. Nevertheless, systematic analysis of the relationships between genomic alterations and transcriptomic alterations is rare. Here, we propose to comprehensively explore the correlations between genomic and transcriptomic alterations. We will correlate the expression level for each gene across all the copy number alterations, somatic mutations and epigenetic modification along the whole genome. We will also investigate whether there are associations between alternative splicing isoform usages and somatic mutations. Using sequencing data from Taiwan oral cancer patients (in-house data) and HNSC (Head-Neck Squamous Cell Carcinoma) from TCGA, we will first establish and optimize the analysis pipeline. The same analysis strategy will also be applied to all the data from different cancer types which will reveal how the interaction networks between genome and transcriptome differ between cancer types. From all these analysis, we can also identify crucial alterations. These alterations are potential biomarkers for cancer detection, prognosis or targets for future cancer drug development. Therefore, we will further exam whether those alterations or the combination of these alterations are associated with clinical factors and test their prediction power in clinical outcomes. After all these analysis, the analysis platform will be released through websites and the same analysis strategy can be easily applied to other genomic studies. In addition to the analysis platform, we will also release the analysis results in an open database. This pan-cancer database will reveal the correlated linkages between genome and transcriptome and provides an invaluable resource for identifying critical alterations in cancers.

Project ID:PA10608-0318
External Project ID:MOST106-2311-B182-005