American Statistical Association
In many recent cancer profiling experiments, samples are being interrogated with microarrays at both a mRNA transcript and a copy number expression level. A crucial question then becomes how to combine the genomic information to better understand the underlying biology as well as to find new candidate biomarkers. In this talk, we discuss our preliminary results in this area. In particular, we find that using simple correlation measures does not capture much of the association between copy number and gene expression. We then describe a hierarchical model for associating the two types of data. It is based on the idea that copy number changes in the genome occur in segments. Bayesian inference is performed in the model using realizations of the posterior distribution, which is sampled using a Markov Chain Monte Carlo-based scheme. Some real datasets are used to illustrate the methodology.
|Date:||Wednesday, May 9, 2007|
|Time:||4:00 P.M. - 5:00 P.M.|
Memorial Sloan-Kettering Cancer Center
Department of Epidemiology and Biostatistics
307 East 63rd Street
(between First and Second Avenues)
3rd Floor Conference Room
New York, New York
Note: To gain access to the building, please follow the directions by the telephone in the foyer.