Title: Detecting single nucleotide variants using feature based classifier
Speaker: Fatemeh Dorri
Department of Computer Science, University of British Columbia
Abstract

Abstract

With the advent of fast and low-cost next-generation sequencing (NGS), the scope for the study of DNA sequences has been increased dramatically. In particular, studying DNA aberrations in oncogenesis have been extensively investigated in order to have a comprehensive understanding about somatic single nucleotide variation (SNV) which is known as one of the key factors playing role in cancer evolution. However, the analysis of DNA sequence data is still challenging to meet high sensitivity and specificity for detecting single nucleotide variations. Here, we propose a tool called Museq that overcomes the above drawbacks for detecting somatic single nucleotide variation. The tool is versatile and efficient while keeping its accuracy and reliability. It can analyze a paired normal-tumor samples considering biological and technical artefacts inherently in the model. In addition, it can detect single nucleotide variations in a single sample and can also be incorporated for deep targeted sequencing data, exon sequencing or whole genome sequencing.