Online supplemental material

Supplemental mateRIalS anD metHODS Microarray processing and data. The manufacturer’s protocols were followed for the determination of gene expression using the GeneChip Mouse Genome 430 2.0 array (Affymetrix). Hybridized arrays were scanned with a scanner (Gene Chip 3000), and image generation and feature extraction were performed using operating software (GeneChip; both from Affymetrix). All arrays passed the manufacturer’s quality specifications with respect to background and percent-present call rates. To generate the demonstrated heat map (Fig. 6 A), a list of 290 differentially expressed probe sets (Table S1) from Socs3 hKO and littermate control mice were identified by the following procedure. Raw microarray data were processed and analyzed with Bioconductor (1) and normalized with the Bioconductor GeneChip Robust Multi-Array Average package (2). After data normalization, genes with significant evidence for differential expression were identified using the limma package (3) in Bioconductor. p-values were calculated with a modified t test in conjunction with an empirical Bayes method to moderate the standard errors of the estimated log-fold changes. p-values were adjusted for multiplicity with the Bioconductor package q value (4). Before adjusting p-values for multiplicity, normalized values were nonspecifically filtered by omitting values whose interquartile range was <0.25. The final list of 290 probe sets (Table S1) was determined by further filtering for fold change > 1.5 and unadjusted P < 0.01 (for which we found that the false discovery rate = 0.084). The overall average intensity for the represented genes in the control littermate arrays was calculated and used as the baseline reference (black) in the demonstrated heatmap (Fig. 6 A). The shades of red or green in each bar on the heat map correspond to an intensity score relative to baseline. Affymetrix accession numbers for probe sets that were differentially regulated in Socs3 h-KO samples were submitted to NIH DAVID for analysis (Dennis, G., Jr., B.T. Sherman, D.A. Hosack, J. Yang, W. Gao, H.C. Lane, and R.A. Lempicki. 2003. Genome Biol. 4:P3). Pathways were categorized by Kyoto Encyclopedia of Genes and Genomes (KEGG) annotation. The percentage of total genes in a given enriched KEGG pathway (Fig. 6 B) was calculated by dividing the number of differentially expressed genes for each pathway identified via DAVID by the total number of genes in the pathway according to KEGG annotation.