Skip to main content

Table 7 Comparisons of study features investigating DNA methylation across the genome of ME/CFS patients compared to controls

From: Changes in DNA methylation profiles of myalgic encephalomyelitis/chronic fatigue syndrome patients reflect systemic dysfunctions

Study

Tissue

Method

Cohort

Diagnostic criteria

Statistical thresholds

No. of significant differences

NZ

PBMCs

Reduced representation bisulphite sequencing

P = 10

5 females

5 males

C = 10

5 females

5 males

Canadian criteria

DMAP ANOVA F test Raw P < 0.05, methylation diff ± 15%

MethylKit Fisher’s exact test FDR corrected P < 0.05. methylation diff ± 15%

DMAP: 76 (52% hypo-methylated)

Methylkit: 394 (56% hypo-methylated)

A

CD4 + T cells

Infinium HumanMethylation450 BeadChip

P = 25

21 females

4 males

C = 18

10 females

8 males

Fukuda criteria

Raw P value < 0.05, methylation fold change > 2.0

120 (85% hypo-methylated)

B

PBMCs

Infinium HumanMethylation450 BeadChip

P = 12, C = 12

All female

Fukuda and Canadian criteria

Wilcoxon-rank sum test P < 0.05, FDR corrected P < 0.05 (Benjamini-Hochberg). Mean beta difference > 0.20

1192 (72% hyper-methylated)

C

PBMCs

Infinium HumanMethylation450 BeadChip

P = 49, C = 25

All female

Fukuda and Canadian criteria

Wilcoxon-rank sum test P < 0.05, FDR corrected P < 0.05. Mean beta difference > 0.05

12,608 (71.6% hyper-methylated)

D*

PBMCs

Illumina Methylation EPIC microarray

P = 13, C = 12

All female

Fukuda and Canadian criteria

FDR-corrected P value < 0.05 absolute beta difference > 0.05

17,296 (98% hypo-methylated)

E

CD3 + T Cells

Infinium HumanMethylation450 BeadChip

P = 43

34 females

9 males

C = 36

27 females

9 males)

Fukuda and Canadian criteria

P value < 0.05 permutation analysis, mean percentage methylation difference > 5%

133 (74% hyper-methylated)

  1. A table comparing the NZ study with the five previous studies investigating DNA methylation in ME/CFS patients vs. controls. (NZ = this study, A = Brenu et al. 2014 [6], B = de Vega et al. 2014 [7], C = de Vega et al. 2017 [8], D = Trivedi et al. 2018 [9], E = Herrera et al. 2018 [10]) The table includes the cell type utilised, the method used in the analysis, the numbers and gender included in the cohort, diagnostic criteria of the patients and the statistical thresholds utilised in the analyses. *D (Trivedi et al. [9]) included a larger cohort for pyrosequencing validation with a total of 33 cases and 31 controls from three geographical locations