Document Type
Lecture
Publication Date
10-29-2004
Abstract
For some of Affymetrix GeneChips, it is necessary to correct the texture effect on the images before doing the common background subtraction and normalization. In this study, we explore a way to assess and correct the texture effect. Without this correction, we found that the texture effect affected the performance of the conventional approach in identification of significantly expressed genes. Furthermore, we found that treating probes as "subjects" in the same probe set and using repeated measures analysis can help remove the local artifacts from the residual images, which are commonly associated with conventional approaches. Finally, we also tried to evaluate and adjust for the effects of several other biological factors on the probe intensity.
Relational Format
presentation
Recommended Citation
Duan, Fenghai, "Probe-level Correction in Analysis of Affymetrix Data" (2004). Probability & Statistics Seminar. 57.
https://egrove.olemiss.edu/math_statistics/57
