GWAS (genome wide association study)

[SNP_HKG]

hello,



Im new to SNP studies and Im looking to clear a few doubts.



There are several studies identifying SNPs near gene loci via genome wide association studies (GWAS). I understand a SNP is significant if it has a genome wide significance of P < 5 × 10−8



If a particular SNP has been identified e.g.

rs2487032 (A), p value 9.20 × 10−14 in disease vs control samples in GWAS


Does this mean that anyone in the population with this SNP

rs2487032 (A) has a high association with that particular disease, if the allele in place is A?

[gringer]

No. Those p-values usually apply to the confidence in the value of the distinguishing statistic (a statistic which, as it happens, is frequently absent from abstracts and the non-supplementary paper text), rather than the association with the compared trait.

When you're reading GWAS results, I recommend you look at those distinguishing statistics (typically beta values), and consider those values when determining whether or not to believe that an association is reasonable.

I don't think the exact p-value should be represented in the abstract or main text of papers, nor as an axis dimension in figures in papers, as it reduces the apparent importance of the distinguishing statistic (which I consider to be more biologically relevant). Many, many people disagree with me on this point.