The cross-sectional CAMP gene expression data have been previously
described (12 (link)). Briefly, 620 CAMP
subjects with whole blood samples collected only at early adulthood were assayed
with the Illumina HT-12 (version 3) platform with 47,009 high-quality probes.
Quality control filters included removal of failed arrays, probes with low
outlying log2 intensities (<5), and probes with poor
signal-to-noise ratios (95th percentile /5th percentile).
These expression data were quantile-normalized and log2-transformed.
A set of 106 subjects who were with obesity at enrollment in CAMP (body mass
index (BMI) ≥ 95th percentile for age and sex) were excluded
to focus the analysis on asthma complicated by incident obesity (10 ). Following the exclusion of probes with low
variances (15 (link)), the final dataset
comprised 514 subjects and 10,448 probes. For replication, cross-sectional whole
blood gene expression data were also assayed with the Illumina HT-12 array
(version 4): 47,009 high-quality probes for 91 adult subjects in CHS (12 (link)); and 47,256 high-quality probes for 329
pediatric subjects in GACRS (16 ). A
subset of 10,448 probes in CHS and 10,437 probes in GACRS were matched to the
CAMP data by their universal nucleotide identifiers (17 (link)).