Abstract

Application of a a random regression model to gene expression profiling  

S. L. Rodriguez-Zas, J. J. Loor, J. K. Drackley, and H. A. Lewin.  University of Illinois, Urbana.

ADSA/ASAS/PSA Annual Joint Meeting, St. Louis, July 25-29, 2004

The patterns of gene expression recorded on individuals over a period can be studied using discrete or continuous representations of time. within the latter representation, the profile of expression can be modeled using common (fixed) and individual (random) polynomial coefficients in time. We evaluated the potential of random regression models to describe the fluctuations in the gene transcription levels recorded at successive time points. The data consisted of fluorescence intensities on more than 6000 unique genes recorded using spotted cDNA microarray technology. Liver samples were obtained at -65 d, -30d, -14d, +1d, +14d, +28d and +49d relative to calving on 8 Holstein cows. A reference design was implemented with each cow-day sample represented in two reverse-dye microarrays and each gene double spotted on each microarray. Fluorescence intensity measurements on 106 microarrays were filtered for weak signals and were normalized using a log2 transformation on the loess-adjusted values. The random regression model included linear to quartic polynomials on days and accounted for heteroscedasticity between days. Three percent of the genes had at least one significant (P < 0.0001) regression coefficient in days. The majority of these genes had significant quadratic trends alone or in combination with a significant quartic trend. Hierarchical and disjoint clustering of these coefficient estimates indicated the presence of 5 clusters. Four of these clusters were approximately characterized by significant (positive and negative) quadratic regression coefficients in combination with significant (positive and negative) quartic regression coefficient within each signed quadratic group. The last cluster was characterized by significant linear and cubic regression coefficients. Results from this study indicate that random regression models are flexible to accommodate variation in patterns that can be observed in genomic studies. Journal of Dairy Science, 87(Suppl. 1):365.