Table 3.
Gene-based high-dimensional mediation analysis for the association between PM10 (exposure), DNAm (mediator) and neurodevelopment (outcome) using gHMA
| Closest gene | CpG count | Cognitive | General adaptive behavior | Language | Motor | ||||
|---|---|---|---|---|---|---|---|---|---|
| Raw p-value | FDR | Raw p-value | FDR | Raw P-value | FDR | Raw P-value | FDR | ||
| A. Significant mediation for the primary outcome (cognitive domain) and cross-validation across domains | |||||||||
| GOPC | 15 | 1.19e−09 | 4.12e−05 | 9.73e−03 | 0.84 | 8.57e−03 | 0.98 | 3.59e−06 | 0.04 |
| RP11-74K11.1 | 20 | 1.13e−07 | 1.95e−03 | 1.13e−07 | 3.91E−3 | 1.05e−07 | 3.6E−3 | 5.48e−02 | 0.89 |
| DYRK1A | 19 | 1.21e−06 | 1.39e−02 | 1.20e−06 | 0.02 | 1.49e−06 | 0.03 | 3.78e−03 | 0.66 |
| RNMT | 8 | 2.86e−06 | 2.47e−02 | 2.87e−06 | 0.03 | 8.94e−06 | 0.09 | 1.22e−02 | 0.83 |
| B. Significant mediation for the secondary outcomes (other domains) and cross-validation across domains | |||||||||
| DCAF13 | 7 | 8.92e−05 | 1.60e−01 | 8.86e−05 | 0.17 | 3.65e−04 | 0.34 | 6.52e−06 | 0.04 |
| TNN | 13 | 4.44e−04 | 3.49e−01 | 4.40e−04 | 0.34 | 1.86e−04 | 0.31 | 8.11e−07 | 0.03 |
| TAL1 | 26 | 4.61e−03 | 8.13e−01 | 1.72e−03 | 0.58 | 9.60e−03 | 1.00 | 4.06e−06 | 0.04 |
| AC011648.1 | 3 | 1.48e−02 | 1.00e+00 | 1.49e−02 | 0.92 | 1.05e−02 | 1.00 | 5.76e−06 | 0.04 |
| SPINK2 | 5 | 9.85e−02 | 1.00e+00 | 4.14e−01 | 1.00 | 4.52e−01 | 1.00 | 6.62e−06 | 0.04 |
Presented associations were significant for at least one neurodevelopmental domain (BH FDR < 0.05). FDR values with bold values indicate significance at a threshold of 0.05