A comprehensive in-silico computational analysis of twenty cancer exome datasets and identification of associated somatic variants reveals potential molecular markers for detection of varied cancer types

Abstract

There are several diseases that are associated with mutations in genes. Cancer is one such disease and is often thought of as challenging to treat in the later stages. Identification of cancer-causing genes in the early stages provides better chances of survival. Therefore, the current study aimed to computationally study twenty cancer exomes belonging to five cancer types and identify the somatic variants that could point towards cancer prognosis. Twenty exome datasets were retrieved and a raw-data pre-processing check including FastQC check, adaptor trimming, gapped alignment, and refinement was performed to assess their qualities. The assessed exome datasets were then used to call the variants and analyze the mutational profiles. Identification of unique SNPs was carried out for each dataset and their functions were scrutinized to find out potential biomarkers. The outcomes of the study revealed that all twenty exome datasets passed the quality checks and 4181 variants were identified post data processing, filtration, and analysis of variants. A comprehensive analysis of the mutational profiles revealed the number of variants that were tolerated, deleterious, probably, and possible damaging as well as benign. CD82 (Cluster of differentiation 82), found in the human diffuse-type gastric cancer dataset and ANO1 (Anoctamine 1) in intrahepatic cholangiocarcinoma, were found to show good gene expression profiles in various cancer types including thyroid cancer, colorectal cancer, head, neck cancer, stomach cancer, and esophageal cancer for CD82 and colorectal cancer, head and neck cancer, stomach cancer, esophageal cancer, urinary bladder cancer, cancers of the kidneys and lungs for ANO1. A comparative analysis between the two potential markers revealed that CD82 was upregulated in a greater number of cancer types than ANO1. Therefore, the present computational study provides preliminary insights into using these potential biomarkers for the early detection and prognosis of varied cancer types.