Statistical Methods for Analysis of Pooled Sequencing Data

• Activity Number: 337
• Type: Topic Contributed
• Date/Time: Tuesday, August 2, 2011 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Zhi Wei*+
• Companies: New Jersey Institute of Technology
• Keywords: next-generation sequencing ; pooled sequencing

Abstract

Although the development of next-generation sequencing technologies (NGS) in the past few years has made the cost of DNA sequencing plummet dramatically, it is still prohibitively expensive to sequence the complete genomes of hundreds of individuals. Just as in pre-GWAS, re-sequencing small targeted genomic regions is now routinely to harness the massive capacity of NGS. Pooled sequencing is used to identify rare variants in targeted regions of the genome in large populations. However, detection of rare variants from pooled sequencing is more challenging than from individual sequencing. Here we propose a new statistical procedure to detect genomic variants from the output of pooled sequencing data. Simulations and analysis of real data will be used to demonstrate the merits of our approach in comparison with existing methods.