Reverse allele frequency

Question

[panacotaforcota]

Hello!

I have a problem with reverse allele frequency in my GWAS. I compare the allele frequency of SNP from GWAS with reference allele frequency. My results after synchronization allele:

So, I want to understand how I can detect such SNPs to remove them from the analysis. Maybe there are common approaches.

[Nicolas Rosewick]

how did you synchronize ?

[panacotaforcota]

By my own script. Briefly, I change allele frequency as (1-freq) in the reference data if alleles were reverse, for example, in GWAS - A:T, in reference - T:A. There is no mistake.

[Nicolas Rosewick]

could you post your code ? Are you comparing the allele freq to gnomAD ?

[German.M.Demidov]

it could be that the reference is given as "mAF", minor allele frequency. Trying to "reverse" in this case will lead to such picture.

[panacotaforcota]

But X and Y axis from 0 to 1. In MAF case 0.5 is upper limit.

[German.M.Demidov]

Unless one tries to "revert" something...

Not sure if it is the cause, but worth to be checked.

the reference data if alleles were reverse, for example, in GWAS - A:T, in reference - T:A