CpG p value threshold | GWAS | p value | Mean difference | FDR-corrected p value |
---|
0.05: | Autism | 0.54 | 0.004 | 0.54 |
0.01: | Autism | 1.0 × 10−4 | 0.039 | 5 × 10−4 |
0.005 | Autism | 1.9 × 10−3 | 0.031 | 4.75 × 10−3 |
0.001: | Autism | 0.040 | 0.037 | 0.063 |
0.05: | SCDC | 0.735 | 0.007 | 0.73 |
0.01: | SCDC | 0.301 | 0.015 | 0.40 |
0.005 | SCDC | 0.022 | 0.038 | 0.046 |
0.001: | SCDC | 0.023 | 0.065 | 0.046 |
0.05: | Alzheimer’s | 0.343 | 0.008 | 0.853 |
0.01: | Alzheimer’s | 0.710 | 0.003 | 0.853 |
0.005 | Alzheimer’s | 0.853 | − 0.003 | 0.853 |
0.001: | Alzheimer’s | 0.793 | − 0.009 | 0.853 |
- The table provides the results of the enrichment analyses for the top loci. We calculated the difference between the average p values for all the mQTLs mapped to CpGs below a selected threshold in the MWAS (CpG p value threshold) and the mQTLs mapped to CPGs above the threshold in the SCDC MWAS. This value is referred to as “mean difference” in the table. A positive difference suggests and enrichment. We then permuted the results after correcting for various factors, and computed a permuted p value (p value). We then corrected it for multiple testing using FDR correction (FDR-corrected p value). This was done using a GWAS for autism, SCDC, and Alzheimer’s