CpG polymorphism

[patid]

Hi,
I am beginner in methylation studies. Recently I received HumanMethylation 450 BeadChip results that point to significant differences in methylation level between two human populations. Those methylation differences were placed beyond CpG island and covered only single CpG. Methylation difference wasn’t found in neighbouring CpG. It is quite strange for me because I expected that such population differences should be observed in bigger genome area- not in a single CpG. When I checked the sequence in Ensamble I have noticed that in this single CpG- nucleotide G is polymorphic in one of the population. Is anyone notice such situation or maybe have read about problem like this? Is it still a methylation studies or just polymorphism?
Many thanks!
Pat

[PeteH]

Hi Pat,

I think you're right to be suspicious of this result. I know that many studies using the 450k or 27k arrays filter out probes that overlap with known SNPs due to the confounding that you identified between the methylation signal and the polymorphism. See, for example, http://nar.oxfordjournals.org/content/41/7/e90, http://www.landesbioscience.com/jour...12EPI0361R.pdf.

I'm not sure there is a satisfactory resolution to this problem, aside from excluding such probes, unless you also have genotype data on the same sample (i.e. from SNP chips or sequencing).

Cheers,
Pete