Add symptoms of depression because 3-way interaction may due to underlying relationship of BDNF with depression.
Table 1.1. Association of BDNF, CRP, race, and 3-way interaction with Oxo
| B | RAfrAm | C | D | B:RAfrAm | B:C | RAfrAm:C | B:RAfrAm:C | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Oxo | -0.00000062 | -0.036 | * | 0.00037 | 0.00012 | 0.0000019 | *** | 0.00000027 | 0.011 | -0.00000061 | ** |
Table 1.2. Association of BDNF, CRP, race, and 3-way interaction with Oxo
| B | RAfrAm | C | D | B:RAfrAm | B:C | RAfrAm:C | RAfrAm:D | B:RAfrAm:C | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Oxo | -0.0000006 | -0.038 | * | 0.00058 | 0.000054 | 0.0000019 | *** | 0.00000026 | 0.011 | 0.00014 | -0.0000006 | ** |
These results suggest that symptoms of depression (D) do not account for the 3-way interaction of CRP by BDNF by race on oxo-G either as a main effect solely or as a main effect and an interaction with race.
We have GWAS results on about one-third of the low and high CRP groups, and about half of the mid-CRP group (each group has 39 participants).
Lo Mid Hi
1 14 20 13