Please note that we have stopped the regular imports of Gene Expression Omnibus (GEO) data into ArrayExpress. This may not be the latest version of this experiment.
E-GEOD-11180 - Transcription profiling of rat chromosome 4 for skeletal traits at femoral neck
Released on 4 April 2009, last updated on 2 May 2014
Femoral neck bone mineral density and structure candidate gene analysis in Fischer 344 (F344) and Lewis (LEW) rats; Hip fracture is the most devastating osteoporotic fracture type with significant morbidity and mortality. Previously, we identified that the region of 4q21-q41 on chromosome (Chr) 4 uniquely harbors multiple femoral neck quantitative trait loci (QTLs) in inbred Fischer 344 (F344) and Lewis (LEW) rats. In this study we identified the candidate genes for femoral neck density and structure by correlating gene expression in the proximal femur with the femoral neck phenotypes linked to the QTLs on Chr 4. Microarray analysis was performed using RNA extracted from proximal femora of 4-week-old rats from F344, LEW and two other strains. A total of 104 genes in the 4q21-q41 region were differentially expressed (p<0.05) among all strains of rats with a false discovery rate (FDR) less than 10%. These 104 genes were then ranked based on the proportion of variation in femoral neck phenotypes in F2 animals homozygous for a particular strain’s allele at the Chr 4 QTL explained by the expression level of the gene in that strain. A total of 37 genes, including 21 candidate genes and 16 predicted genes, were strongly correlated (r2>0.50) with different femoral neck phenotypes and prioritized for further analysis. Ingenuity pathway analysis revealed several direct or indirect relationships among the candidate genes related to bone metabolism including pathways related to beta-estradiol, interleukin 6, insulin growth factor 2, androgen receptor and tumor necrosis factor. Experiment Overall Design: Comparison of differentially expressed genes at the chromosome 4 QTL identified in F344 and LEW F2 rats.
transcription profiling by array, unknown experiment type