Table 1.
Dementia |
Alzheimer’s disease |
||||
---|---|---|---|---|---|
N CpG sites | Total sample | Co-twin control | Total sample | Co-twin control | |
BIN1 | 59 | 0.07 | 0.69 | 0.05 | 0.58 |
HLA-DRB5 | 12 | 0.02 | <0.01 | 0.04 | 0.18 |
SORL1 | 47 | 0.06 | 0.75 | 0.10 | 0.96 |
SLC24A4 | 48 | 0.01 | 0.17 | 0.01 | 0.16 |
ABCA7 | 202 | 0.24 | – | 0.23 | – |
Number of CpG sites included and gene-level significance for each of the five genes. Logistic regression models were applied to test associations between CpG M-vaules and dementia or Alzheimer’s disease. Models using the total sample were adjusted for age at blood sample, sex, and relatedness among twins, and co-twin control models adjusted for age at blood sample. P-values for each CpG site across the genes were combined into a test statistic using the Fisher product method, and the statistical significance tested through random permutations. Bold numbers indicate significance at the α = 0.05 level. The co-twin control permutation test for ABCA7 did not converge.