Summaries for High Density Oligonucleotide Array Probe Level Data

• Activity Number: 346
• Type: Topic Contributed
• Date/Time: Wednesday, August 14, 2002 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section*
• Author(s): Leslie Cope*+ and Rafael Irizarry+
• Affiliation(s): Johns Hopkins University and Johns Hopkins University

Abstract

In this talk, we report some exploratory analyses of high-density oligonucleotide array data from the Affymetrix GeneChip system. Our analyses make use of four data sets: two small experimental studies, part of the data from an extensive spike-in study, and part of a dilution study. We explain why we need to normalize the arrays to one another and why we believe it should be done at the probe level. We calculate and explore four summary measures of expression: Affymetrix's: I.) average difference (AvDiff); ii.) MAS 5.0/Lims 3.0 Statistical Expression Algorithm signal (MAS 5.0); iii.) the Li and Wong multiplicative model-based expression index (MBEI); and iv.) a measure AvLog (PM-BG), where we use a simplistic approach to calculating a background adjustment BG. We evaluate the four expression summary measures using the dilution study data, assessing their behavior in terms of bias, variance, and for Li and Wong's MBEI and AvLog (PM-BG)) model fit. Finally, we evaluate the algorithms in terms of their ability to detect known levels of differential expression using the spike-in data.