Novel Signal Processing-Based Methods for the Detection of Copy Number Changes in Human DNA

• Activity Number: 581
• Type: Contributed
• Date/Time: Wednesday, August 4, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Catherine Stamoulis*+ and Rebecca Betensky
• Companies: Harvard Medical School and Harvard School of Public Health
• Keywords: genomic data ; signal processing ; copy number variations ; detection

Abstract

Chromosomal copy number changes (CNVs) occur across the human genome as part of our normal genetic heterogeneity, but have also been implicated in a range of diseases and disorders, such as cancer. Although large CNVs may be detectable in noisy genomic data, small-scale localized allelic changes may be undetectable in these data. Available methods do not explicitly target the signal-to-noise ratio (SNR) which may affect significantly the performance of CNV detectors. We adapted signal processing and pattern matching methods to detect common and pathological CNVs in Array Comparative Genomic Hybridization (aCGH) data from healthy donors and meningioma patients. Our methods successfully suppressed platform-independent genomic artifacts and increased the data SNR by at least one order of magnitude in normal and meningioma DNA, thus enabling robust CNV and associated breakpoint detection.