Statistical Analysis of Copy Number Variation with Sequencing Data

• Activity Number: 611
• Type: Contributed
• Date/Time: Thursday, August 7, 2014 : 8:30 AM to 10:20 AM
• Sponsor: Section on Medical Devices and Diagnostics
• Author(s): Bipasa Biswas*+ and Yinglei Lai
• Companies: CDRH/FDA and George Washington University
• Keywords: Next Generation Sequencing ; Genomics ; Copy Number Variation ; Finite Mixture Models ; Mixture of Generalized Linear Models ; Geometric Distribution

Abstract

Copy Number Variation (CNV) has been explored in recent years using computational and statistical methods based on the count type data collected by either a fixed or a sliding window approach. We propose to explore CNV using distance data where distance is measured in base pairs (bps) between two adjacent alignments of short reads mapped to a reference sequence. The advantage of our approach is that we do not need a moving window and additionally the bin size does not need to be defined, thus all local information along the chromosome is not lost. Secondly, we used an user defined censoring at distance 5000 bps for possible chromosomal deletions which is useful when comparing abnormal chromosome to its paired normal. We propose a mixture of generalized linear models (GLMs) extended to mixture of right censored geometric distributions to model the distance data which arise from next generation DNA sequencing instead of count. The model was developed based on the Newton-Raphson algorithm as well as the Expectation-Maximization algorithm. We applied our method to the sequencing data set where we picked a normal and a cancer cell line for a change-points analysis. The rank based inver