Analyses of gene expression using combined ovine and bovine transcriptional profiling for identification of common lactation pathways M. Singh, P.C. Thomson, P. Sheehy and H.W. Raadsma Centre for Advanced Technologies in Animal Genetics and Reproduction (ReproGen), Faculty of Veterinary Science, University of Sydney, PMB 3 (425 Werombi Road), Camden NSW 2570, Australia A number of differentially expressed (DE) candidate genes have been proposed from expression levels obtained in individual microarray experiments for a number of species. The combination of information obtained from such datasets can lead to a list of differentially expressed genes that signify unique and conserved molecular processes that are functionally important. Such models are potentially more robust with greater confidence, and place less reliance on a single dataset. In this paper we combine two microarray datasets for three stages of lactation in both sheep and cattle and perform comparisons of heterogeneous data. An Affymetrix bovine gene chip array containing about 24,000 probe sets was hybridized with RNA derived from mammary gland of pre, peak and late lactation stages in dairy sheep selected for high and low yield milk traits, providing a set of 12 chips. Data from an additional study using the same Affymetrix chip and experimental design, with 15 bovine biological replicates over corresponding pregnancy, lactation and involution stages was used. The combined data were corrected for background noise and normalised using the RMA normalisation method. Gene effects were estimated using a mixed model approach and a mixture model distribution was then fitted to the sets of estimated effects to obtain posterior probabilities for identification of differentially expressed genes. Clusters of top 20% of these genes for the two livestock species will show the DE genes that are co expressed in both representing conserved differential expression in the target mammary gland tissue. Further analysis on exploring the contrasts and similarities between the DE genes of these two species will be explored by analysing the two datasets separately using the same quality control, normalisation and analysis processes. Gene expression pathway analysis results of the two studies can also be compared to identify common clusters of genes representing the same functions across both species. Analysis of Microarray Gene Expression Profiles In Mammary Glands of CBA and QSI5 Mice Jerry Wei, Peter Thomson, , Palaniappan Ramanathan, and Peter Williamson 1 Centre for Advanced Technologies in Animal Genetics and Reproduction (ReproGen), Faculty of Veterinary Science, 2 CRC for Innovative Dairy Products, University of Sydney The use of mouse models for mammary gland research has generated a number of discreet datasets. Comparing strains of mice, is also considered an effective approach for identifying key candidate genes. The QSi5 strain, developed at The University of Sydney, has demonstrated a superior lactation performance when compared to other inbred mouse strains. Comparison of gene expression profiles in this strain with other strains will help in identifying the genes responsible for the lactation performance phenotype, but this will be most effective using methods that permit inter-study comparison. Inguinal mammary glands were harvested from 5 animals in each of 2 strains at Day 12 of pregnancy. RNA isolation was performed using standard protocols and were hybridised to mouse Affymetrix gene expression arrays (430). A mixed model method of analysis was used to compare expression profiles of these microarrays and to compare to the Limma method. The analysis algorithm was modified to Poster Abstracts