FIG. 1.

Unsupervised clustering of monocyte gene expression profiles identifies a subgroup of recent-onset type 1 diabetic patients. Purified PB monocytes from patients (D) or healthy control subjects (HC) (H) were expression-profiled by Illumina whole genome microarray, in two independent experiments (A, B). For each microarray, all genes detected in ≥1 sample were clustered using the Euclidean Distance correlation. Monocyte expression profiles from all patients, or the subset of patients designated “B”, were used to identify genes that were significantly differentially expressed (≥1.5-fold differential expression, Benjamini Hochberg corrected P < 0.05) between patients and control subjects (C, D).