Effect allele in TrajGWAS when using a PLINK file

[hmtsao]

Hi,
I ran a TrajGWAS analysis using a PLINK file, score test first then Wald test for the selected variants. The analyses went smoothly. However, the beta values in my result were positive, which was not what I expected.
In the PLINK file, the A1 allele is set as the effect allele and the A2 is the reference allele. I wonder if TrajGWAS uses the same setting as PLINK or the other way around.

Thanks for all the help.

[kose-y]

@hmtsao TrajGWAS uses A2 as the effect allele. In fact, there is no standard for ref/alt designation in the PLINK 1 format, so it depends on project-wise convention. We will consider adding ability to flip ref/alt designation for the PLINK input.