I am reading a GWAS paper that found a SNP associated to predisposition to colon cancer and was assessed for gene expression of the nearby gene. They found that the genotype accounted for 55% of the variation in the nearby gene expression. 55% sounds like a lot to me, what is the usual percent accounted for a significant SNP in a study like this?
eQTLs (expression quantitative trait loci) are variants that affect the expression of one or more genes.
There have been several 'genome-wide' studies of SNPs that directly affect expression. The actual effect sizes are hard to pin down in most of them, but in the supplementary data for this paper is a list of the SNPs with the largest effects and coefficients of determination.
There are several common SNPs (cis-acting) that account for >50% of the variation in that genes expression in this (European) population cohort. That being said, that vast majority of variants account for much less variation in the expression, but this is to be expected.
I therefore can fully suppose that a disease-linked SNP could affect the expression of a gene to that degree, particularly if it is an oncogene or tumour suppressor.
- Wood AR, Hernandez DG, Nalls MA, Yaghootkar H, Gibbs JR, Harries LW, Chong S, Moore M, Weedon MN, Guralnik JM, et al.. 2011. Allelic heterogeneity and more detailed analyses of known loci explain additional phenotypic variation and reveal complex patterns of association. Human molecular genetics 20: 4082–92.