Analysis of Copy Number Variation in Esophageal Cancer

ABSTRACT There is an immense amount of variability in the human genome that is caused by structural variation. Of these categories of variation in the genome, Copy Number Variation (CNV) is a prevalent form of structural variation where, when compared to a reference genome, alterations in the DNA give rise to a deviation in the number of copies of one or more multi-nucleotide segments of DNA. These variations, dependent on the location, can affect gene expression and it has been suggested that CNV could play a significant role in the development of cancer. An over expression of an oncogene or under expression of a tumor suppressor gene could promote the development and occurrence rates of an associated cancer. Esophageal squamous cell carcinoma (ESCC) is among the most common and deadliest cancers, ranking the eight most frequently diagnosed cancer globally and the sixth most in many asian countries(1). ESCC is a highly aggressive cancer, and due to its proclivity to metastasize and invade surrounding tissues, has a remarkably low survival rate. This study investigates techniques of identifying Copy Number Variation in tumor normal pairs collected from patients with esophageal cancer, and uses these results to identify genes that are commonly associated with the disease. Our own custom method for CNV detection was developed and run alongside an established industry standard implementation. Our method performed agreeably against our verification data as well as against the production tool on NGS data, reporting notably similar CNV locations per exon. Furthermore, we were able to identify a promising set of genes of interest by overlapping common CNVs per sample.