Prediction of causative variants in cancer using NGS and GWAS

Abstract

The Cancer Genome Atlas (TCGA) data is used for analysis in this study for prediction of causative variants in cancer. We have used Next-Generation Se quencing (NGS) data containing copy number variants and Genome Wide Associ ation Studies (GWAS) data containing single nucleotide variants in three different types of lung related cancers (Lung adenocarcinoma, Lung squamous cell carci noma and Mesothelioma) using the customised pipelines. Annotation (gene based, transcription factor binding sites, conserved elements, microRNAs and snoRNAs), functional enrichment analysis and protein-protein interaction have been covered in this study. Variants lying in highly conserved regions or overlapping the highly conserved regions are identified. Our results show that three types of microRNAs (hsa-mir-3149, hsa-mir-933 and hsa-mir-4307) were common in all three types of Lung cancers. Genes containing zinc finger domain were identified in all three type of lung cancers. Transcription factor binding sites (nks39, cdc5 and foxo3) were common in all three types of cancers, suggesting their regulatory function.