Addressing Trait-Dependent Expression of Genes in QTL-Regions for WHC By Tiling Array and mRNA-Seq Analysis

Interactions among genes, transcripts, proteins, metabolites and with environmental factors finally determine the organismal phenotype. Variability in meat quality traits like water holding capacity (WHC) is a major concern for producers and consumers. In order to detect relevant molecular pathways and genes, we integrated genome-wide SNP genotypes and holistic data of mRNA and miRNA abundances. As a complementary approach we constructed a 1M feature tiling array targeting previously detected QTL and eQTL-regions. The tiling array covers 18 Mb of genomic sequences comprising of 254 annotated transcripts and representing 234 genes. The array also covers 113 genes that are not represented by the Affymetrix GeneChips previously used for the eQTL study and regulatory network construction. The expression values of common genes on both types of arrays were significantly correlated (r = 0.6). Comparisons of RNA samples between high- and low-WHC animals revealed 37 and 30 genes with different transcript abundances in a Duroc × Pietrain F2- and a Pietrain × (German Landrace × Pietrain) population, respectively. Of these, 14 transcripts were commonly found in both populations, and they were not previously detected using the catalogue arrays. A comparison between tiling array results and mRNA-seq counts of the corresponding transcribed genomic regions revealed a high concordance. Real-time PCR validation of selected transcripts showed significant correlations of 0.5 - 0.8 and 0.6 - 0.7 for the tiling array and the mRNA-seq analyses, respectively. Integration of genomic data at the genome and transcriptome levels by various methods which complement each other can provide multiple evidences of gene candidates for complex traits.