|
Organizers |
Mixed model analysis of contributions to transcriptional variance
by
Greg Gibson
Dept of Genetics, North Carolina State University
Coauthors: Russ Wolfinger (SAS Institute)
While most practitioners of microarray and GeneChip analysis of transcription have focused on the identification of subsets of genes that show significantly different expression between treatments, there are many applications for which partitioning of the variance components is of equal importance. We have developed a mixed model approach, implemented in SAS, that allows us to assess the significance and magnitude of fixed effects such as sex, age, drug, and strain, as well as interactions among these, on global patterns of gene expression. Application to data from the fruitfly, Drosophila melanogaster, suggests that well over half of the genome is differentially expressed between the sexes, and that up to one quarter of the genome may be differentially expressed between wild type strains. Issues relating to power and interpretation of analyses will be discussed.
Date received: July 30, 2001
Copyright © 2001 by the author(s). The author(s) of this document and the organizers of the conference have granted their consent to include this abstract in Atlas Conferences Inc. Document # cahg-10.