Help with LD interpretation of clinically relevent SNPs

Question

[jarrodheck]

Greetings,

I am trying to interpret the following paragraph from the CPIC: Clinical Pharmacogenetics Implementation Consortium that outlines the LD for the following SNPs:

rs12248560 (17) and rs4244285 (2)

The excerpt states that the two SNPs have a D'=1 and an r^2=0.06ish depending on the HapMap population. What confuses me as these two SNPs are well known to exist in isolation. The 17 variant does not always occur in the 2 background, as my interpretation of the excerpt would suggest. Is the CPIC guideline wrong, or am I not understanding something about D' and r^2? Or could there be some population specific effect I am not considering?

The defining polymorphisms of CYP2C192 and 17 are c.681G>A and c.-806C>T, respectively. There is linkage disequilibrium (LD) between c.681G and c.-806T (e.g., |D'|=1.0 and r2=0.064 in CEU HapMap sample; |D'|=1.0 and r2=0.065 in YRI HapMap sample; and |D'|=1.0 and r2=0.074 in CHB HapMap sample). This means that the less common *17 variant (c.-806T) always tracks on the same allele with the more common c.681G. This complicates any interpretation of whether these two variants act independently of one another, and published articles argue both for (29, 35) and against (15, 36, 37) this point.

Thanks for any input! :)

Jarrod

Answer 1

[Fabio Marroni]

D'=1 probably means that you only see three of the four possible haplotypes. Actually, the excerpt states:

"This means that the less common *17 variant (c.-806T) always tracks on the same allele with the more common c.681G"

So, basically you only see these three haplotypes

• c.-806T/c.681G
• c.-806C/c.681G
• c.-806C/c.681A