Hi Kyoko,
I hope you are doing okay. :)
I just wanted to ask a question RE: the difference between Lead SNPs and Ind Sig SNPs. I thought I had grasped the difference but reading your Nat Genet paper on Alzheimer's had me doubting myself
From what I understand this is the following criteria for lead SNPs vs Ind Sig SNPs (using default FUMA settings)
Ind Sig SNPs:
- Association P Val <=5e-8
- R2 <0.6 (independent of each other)
Lead SNPs:
A subset of Ind sig SNPs, therefore following criteria is the same
- Association P Val <=5e-8
- R2 <0.6 (independent of each other)
In addition:
- the second clumping: they have an R2 >0.1 with each other?
However, in the Jansen Alz paper (10.1038/s41588-018-0311-9), it mentions that lead snps are a subset of ind sig snps with R2>0.6 - I don't understand how that can be the case when all lead SNPs are independent of each other (have an R2 >0.1, but <0.6)? Have I confused things?
Genomic risk loci definition. We used FUMA26 v1.2.8, an online platform for
functional mapping and annotation of genetic variants, to define genomic risk loci and obtain functional information of relevant SNPs in these loci. We first identified independent significant SNPs that have a genome-wide significant P value (< 5 × 10−8
) and are independent of each other at r2 < 0.6. These SNPs were further represented by lead SNPs, which are a subset of the independent significant SNPs that are in approximate linkage equilibrium with each other at r2 > 0.6. We then defined associated genomic risk loci by merging any physically overlapping lead SNPs (LD blocks < 250 kb apart). LD information was calculated using the UKB genotype data as a reference.
Thank you!
Waheed