Cross-sample Modeling of Genome-wide Profiling Data, with Examples in DNA Copy Number Analyis

• Activity Number: 218
• Type: Invited
• Date/Time: Monday, August 2, 2010 : 2:00 PM to 3:50 PM
• Sponsor: International Chinese Statistical Association
• Author(s): Nancy Zhang*+
• Companies: Stanford University

Abstract

With the advent of high-throughput data collection schemes in biology, there is now an abundance of data sets where a linearly ordered sequence of observations is collected in parallel over many samples. For example, the data may be measurements of DNA copy number, protein binding, or RNA expression, mapping to ordered locations along a genome. Both the length of each sequence and the number of samples may be very large. Given this data, one is often interested in jointly segmenting all sequences simultaneously to give a sparse cross-sample summary for downstream analysis. I will describe change-point models for this problem and, in particular, focus on model selection procedures to determine the location and carriers of the signals.