Implications of Proteinchip® Surface-enhanced Laser Desorption/Ionization (SELDI) Technology on Statistical Techniques Used in Proteomics.

• Activity Number: 387
• Type: Contributed
• Date/Time: Thursday, August 15, 2002 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section*
• Author(s): Sreelatha Meleth*+ and Mark Carpenter and Dung-Tsa Chen
• Affiliation(s): University of Alabama, Birmingham and University of Alabama, Birmingham and University of Alabama, Birmingham
• Keywords: proteomics ; 2d-gel ; multivariate analysis ; protein array

Abstract

The advent of the Proteinchip® surface enhanced laser desorption/ionization (SELDI) technology is considered a major improvement on existing 2D gel electrophoresis. While traditional proteomics has already developed statistical techniques to compare protein expression levels between groups, the Proteinchip® ability to detect trace proteins is expected to produce expression data many fold greater in dimension. These chips will produce several thousand protein expression values for any given tissue sample. Traditionally, statistical analysis of proteomics data has used multivariate techniques, such as Principal Component Analysis, Partial Least Squares Regression, and Discrete Wavelet theory to reduce dimensionality in the data and to discriminate between clusters. However, when experiments provide very few observations, but thousands of protein expression values per sample, these techniques should be applied cautiously. In this paper, we will review applicability of current techniques in proteomics to ProteinChip array data, as well as recently developed statistical methods in microarray cDNA & mRNA expression data analysis to ProteinChip expression data.