Table 2. Differentially methylated probes for each cell population in comparison to whole blood.
M-value comparison | Comparison on calls* | Absolute variation in comparison to whole blood (M-values)** | Unmethylated | Marginal | Methylated | |
PBMC | 23226 | 5693 | 1.2–11.1 | 1.7% | 25.8% | 72.5% |
CD4+ T-cells | 58980 | 18537 | 1.2–51.9 | 9.4% | 34.5% | 56.1% |
CD8+ T-cells | 125713 | 32948 | 1.2–93.2 | 6.8% | 59.7% | 33.4% |
CD56+ NK cells | 92238 | 23484 | 1.1–54.0 | 6.5% | 58.4% | 35.1% |
CD19+ B cells | 97377 | 23245 | 1.2–221.8 | 15.0% | 55.7% | 29.3% |
CD14+ Monocytes | 41063 | 12625 | 1.2–70.4 | 71.2% | 13.2% | 15.6% |
Granulocytes | 27365 | 8815 | 1.2–13.1 | 85.3% | 9.5% | 5.2% |
Neutrophils | 28581 | 10808 | 1.2–14.5 | 84.2% | 10.1% | 5.7% |
Eosinophils | 52431 | 17723 | 1.2–117.2 | 65.8% | 23.3% | 10.8% |
PBMC-Peripheral blood mononuclear cells. Differentially methylated probes were defined by a linear model using the M-values. M-value is the log2 ratio of the intensities of methylated probe versus unmethylated probe, a measurement of how much more a probe is methylated compared to unmethylated [46]. *To extract the probes with largest difference in methylation, a gamma fit model was applied to M-values in order to define the three calls: “unmethylated”, “marginal” and “methylated”. Significant probes sharing the same call in the two compared populations were removed. **Variation is based on the estimate of the log2-fold-change corresponding to the effect obtained from the linear model, absolute M-values are presented. The percentages are based on the call distribution.