Franny Jongbloed

99 4 A PROTEIN-FREE DIET PROTECTS AGAINST RENAL IRI number of DEPS in experimental group B, nC the number of total genes in the microarray, and nAB the number of commonDEPS between nA and nB. Data integration of microarrays hybridized in different dates was performed with meta-analysis. The methodology applied was combining rank orders. It is a non-parametric approach based on rank orders. The R package RankProd implemented in INMEX was used 20 . In summary, for each dataset, the fold changes (FC) were calculated for all possible pairwise comparisons. The ranks of the fold changes within each comparison were used to calculate the rank product for each gene. To assess the null distribution of the rank product within each data set, a permutation test was performed. The process was repeated several times to compute the P- value and false discovery rate (FDR) associated with each gene. A gene was selected as differentially expressed if it hadanFDR<5%andanabsolute combinedFC≥1.5. Functional annotationand analyses were performed using Ingenuity software (http://www.ingenuity.com/ products/ ipa). The prediction inhibition or activation of the upstream transcription regulators is calculated via de statistical z-score based on the observed gene expression changes in our dataset. Calculating the z-score reduces the chance of significant predictions based on random data (http://ingenuity.force.com/ipa/articles/FeatureDescription/Upstream -Regulator-Analysis). Cutoff values for a significant activation or inhibition were met with a z-score of ≥ 2 or ≤-2, respectively.