American Statistical Association
Chromosomal aberrations are critical regulators of the neoplastic phenotype. Pathogenic alteration in allele dosage facilitates the functional effect of oncogenes and tumor suppressors in disease genesis and progression. Concomitantly, high-density array-based methodologies are rapidly identifying loci of DNA copy-number change. Nevertheless, deciphering from this bounty of highly dimensional data potentially causal somatic changes from those biologically tolerated passenger alterations that are simply the consequence of disease evolution poses a unique analytical challenge. Here we describe a systematic approach, RAE, for the accurate statistical assessment of chromosomal alteration in tumor copy-number data. We discuss our approach to the attenuation of multiple heterogeneous sources of noise, the use of a multi-component system of soft discriminators for single-sample detection, the derivation of a common breakpoint profile for whole-cohort evaluation, the generation of a biologically realistic background model, and the propagation of intrinsic error to the problem of identifying regions of interest (ROI). We demonstrate the power of integrating this robust statistical methodology with high-resolution genomic data to comprehensively catalogue alteration in the soft-tissue sarcoma genome.
|Date:||Wednesday, January 23, 2008|
|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.