Total RNA was isolated from whole blood using the PAXgene® Blood RNA Kit 50 (Qiagen ref.762174). The total RNA was cleaned and concentrated to obtain smaller elution volumes using RNeasy® MinElute (Qiagen cat 74204).
Label
biotin
Label protocol
Standard Affymetrix protocol
Hybridization protocol
Standard Affymetrix protocol
Scan protocol
Arrays were scanned using GeneChip ® Scanner 3000 (Affymetrix). Image analysis was performed using GeneChip ® Operating Software 1.4 (Affymetrix)
Description
treatment group 2
Data processing
The data were normalized using the farms package (Hochreiter 2006) in R (version 2.10.1), using the RMA (Irizarry 2003) method. Probes were summarized using the revised entrez-based probe annotation of Dai (Dai 2005, the package hgu133plus2hsentrezgcdf.db, version 12.1.0), and classified as informative or non-informative using the method of Talloen (Talloen 2007). Non-informative probe sets were discarded. The latter two methods reduced the number of probe sets from 54,675 on the original chip, to 17,788 after re-annotation, and to 3,640 after discarding non-informative probe sets (filtered data provided as a supplementary file on GSE24859). Differentially expressed probe sets were identified using the limma R package (Smyth 2004) by subtracting baseline measurement for each patient from the measurement at six weeks, and compare the resulting expression differences between patients with different treatments using limma's moderated t-test. P-values were corrected for multiple testing using the false discovery rate-controlling procedure of Benjamini and Hochberg (Benjamini 1995).