Hello,

I have found an association signal about 300kb away from a known risk locus.

What I would like to do is determine that the new association is independent of the old, and therefore that it is a novel independent risk locus.

Unfortunately, I only have association summary statistics, not original genotypes I am familiar with the procedure for conditioning the association of one SNP on the other if I have genotypes, but I am less certain how to proceed if all I have is the summary stats.

I can approximate LD between the variants using reference panels such as 1000 genomes data... using this approach, I checked linkage between the two index variants - it is low; r^2 < 0.01 in all subpopulations of interest.

I think there is a way to condition association summary statistics on one another provided one has LD info, but I would really like to know if there is an established method for making this type of claim. Thank you.