Have questions? Visit https://www.reddit.com/r/SNPedia

Talk:Rs9886784

From SNPedia

After reading over [PMID 17998437], from what I can tell it seems rs9886784's association with Alzheimer's disease is found only under a recessive model affecting the (A;A) genotype, and was not found to produce any significant effect for (A;C) heterozygotes—something which can also be inferred given that the percentage of both of the data sets' heterozygote Alzheimer's patients hardly differs from that of either of the heterozygote control groups (.306 vs .313) & (.316 vs .306) for the Canadian and UK data sets, respectively. See the table here: (https://i.imgur.com/TweT4SC.png).

So given the (A;C) genotype's likely benign status and its significant prevalence within the population, it may be prudent to update its genotype page's details to match those of the normal (C;C) genotype, and perhaps change the (A;A) genotype details from the current somewhat imprecise and alarming ">3.23 x risk for AD", to a more accurate variation using the results of the two data sets found in the study i.e.(2.90-3.23 times Increased risk for Alzheimer's). Lastly, given this SNP's failure to replicate in [PMID 19204163], it may also be a good idea to lower the magnitude a bit for now pending further study. --Neverknowsbest (talk) 22:15, 18 July 2017 (UTC)

eminently reasonable. I've knocked down the genotypes, but would encourage you to make further edits directly. --- cariaso 20:55, 19 July 2017 (UTC)