Using immunomagnetic beads (Dynabeads, Invitrogen, Carlsbad, CA), CD14+ monocytes as well as CD34+ stem cells were positively isolated for direct cell lysis, while negatively isolated monocytes were split into two fractions for stimulation with 10 ng/ml lipopolysaccharide (LPS) for 3h, or for 20h cell culture towards macrophages.
Extracted molecule
total RNA
Extraction protocol
Positively isolated monocytes and stem cells as well as cultured stimulated monocytes and macrophages were lysed and total RNA was isolated (Absolutely RNA Microprep Kit, Stratagene, La Jolla, CA)
Label
biotin, Cy3
Label protocol
Total RNA samples fwere amplified and biotinylated using the Illumina TotalPrep RNA amplification Kit (Ambion, Austin, TX).
Hybridization protocol
According to beadchip array manufacturer's protocol
Scan protocol
According to beadchip array manufacturer's protocol
Description
1740115122_C
Data processing
Array data were extracted using Illumina's BeadStudio software. One mislabeled array and ten low-signal arrays (corresponding to seven unique samples and one triplicate sample) with less than 30% of the probes having a detection p-value <0.01 were removed, leaving a total of 151 arrays for analysis (including 39 technical replicates). From the CD34+ cell samples, 38 arrays were analyzed, including 8 technical replicates. Normalization and statistical analysis of the bead summary data from the arrays was carried out using the limma package14 and in-house scripts in R/Bioconductor. Bead summary intensities were log2-transformed and then normalized using quantile normalization. To find differentially expressed genes, we performed a linear model analysis. Technical replicates were handled by estimating a common value for the intra-replicate correlation and including it in the linear model. Differential expression between the treatments of interest was assessed using a moderated t-test. This test is similar to a standard t-test for each probe except that the standard errors are moderated across genes to ensure more stable inference for each gene. Resulting p-values were corrected for multiple testing using the Benjamini-Hochberg false discovery rate.
