GWAS: How to estimate probability of SNP association being result of one specific regulatory element, rather than others in LD
1
0
Entering edit mode
7.3 years ago
michael.nagle ▴ 100

I have GWAS data indicating a large number of SNPs may be associated with a trait of interest. Some of these SNPs are within or very nearby regulatory elements, such as long ncRNAs.

These SNPs are markers for alleles from many different genotypes. It is unclear whether the regulatory elements near these SNPs are the cause of the associations with the trait – or if another element within LD is causing the association

I'm trying to figure out a way to calculate the probability of these lncRNAs being the cause of the associations. I don't yet know how to do this, but have been thinking of an approach using integral calculus and the following variables:

Variables I know how to incorporate:
- rate of LD decay
- length of regulatory element
- position of SNP relative to the regulatory element of interest

Variables I don't know how to account for:
- strength of association
- nearby associated SNPs, also within LD, and the strengths of these associations
- variations in LD dependent on local genomic structure

The approach I'm considering would depend on calculating the black area in the figure relative to the blue area. I think this approach may be rather crude and am wondering about ways to make the estimation more accurate, by including more variables such as the ones I listed and others.

Figure demonstrating the crude approach I am currently trying to improve

GWAS SNP LD Algorithms Calculus • 2.1k views
ADD COMMENT
0
Entering edit mode
7.3 years ago
Sam ★ 4.8k

I'd suggest you to perform a partitioning of heritability and see if majority of the heritability lies within these regulatory elements. LDSC is one of the tool that can perform that

Though the problem of this approach is that it doesn't provide you with a SNP based probability

ADD COMMENT

Login before adding your answer.

Traffic: 1553 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6