Research Article

Detecting the potential cancer association or metastasis by multi-omics data analysis

Published: August 19, 2016
Genet. Mol. Res. 15(3): gmr8987 DOI: https://doi.org/10.4238/gmr.15038987
Cite this Article:
(2016). Detecting the potential cancer association or metastasis by multi-omics data analysis. Genet. Mol. Res. 15(3): gmr8987. https://doi.org/10.4238/gmr.15038987
646 views

Abstract

Comprehensive multi-omics data analyses have become an important means for understanding cancer incidence and progression largely driven by the availability of high-throughput sequencing technologies for genomes, proteomes, and transcriptomes. However, how tumor cells from the site of origin of the cancer begin to grow in other sites of the body is very poorly understood. In order to examine potential connections between different cancers and to gain an insight into the metastatic process, we conducted a multi-omics data analysis using data deposited in The Cancer Genome Atlas database. By combining somatic mutation data along with DNA methylation level and gene expression level data, we applied a Bayesian network analysis to detect the potential association among four distinct cancer types namely, Head and neck squamous cell carcinoma (Hnsc), Lung adenocarcinoma (Luad), Lung squamous cell carcinoma (Lusc), and Skin cutaneous melanoma (Skcm). Further validation based on the ‘identification of somatic signatures’ and the ‘association rules analysis’ confirmed these associations. Previous investigations have suggested that common risk factors and molecular abnormalities in cell-cycle regulation and signal transduction predominate among these cancers. This evidence indicates that our study provides a rational analysis and hopefully will help shed light on the links between different cancers and metastasis as a whole.

Comprehensive multi-omics data analyses have become an important means for understanding cancer incidence and progression largely driven by the availability of high-throughput sequencing technologies for genomes, proteomes, and transcriptomes. However, how tumor cells from the site of origin of the cancer begin to grow in other sites of the body is very poorly understood. In order to examine potential connections between different cancers and to gain an insight into the metastatic process, we conducted a multi-omics data analysis using data deposited in The Cancer Genome Atlas database. By combining somatic mutation data along with DNA methylation level and gene expression level data, we applied a Bayesian network analysis to detect the potential association among four distinct cancer types namely, Head and neck squamous cell carcinoma (Hnsc), Lung adenocarcinoma (Luad), Lung squamous cell carcinoma (Lusc), and Skin cutaneous melanoma (Skcm). Further validation based on the ‘identification of somatic signatures’ and the ‘association rules analysis’ confirmed these associations. Previous investigations have suggested that common risk factors and molecular abnormalities in cell-cycle regulation and signal transduction predominate among these cancers. This evidence indicates that our study provides a rational analysis and hopefully will help shed light on the links between different cancers and metastasis as a whole.

About the Authors