multiple testing is a bitch, huh?

Thank you!

Another important factor is where on the phylogeny (or genealogy) the mutations occur. It is very plausible to get a false signal because the closest SNP occurred on the wrong branch, but a more distant SNP is in a very convenient position.

multiple testing is a bitch, huh?

yeah, but I'm an optimist-- new methods and designs will find a way to minimize the problem.

Another important factor is where on the phylogeny (or genealogy) the mutations occur. It is very plausible to get a false signal because the closest SNP occurred on the wrong branch, but a more distant SNP is in a very convenient position.

I wouldn't call that necessarily a false signal-- assuming we're talking about association, the hypothetical distant SNP would have to be in strong LD with the causal allele (as measured by r2) in order for the association to be detected. so it's not likely to be that far away (though in theory it could be anywhere in the genome).

JP -- having problems with that link -- can you re-point? Thanks

http:// www.journals.uchicago.edu...5604452519Guest

sorry

http://www.journals.uchicago.edu...0012/ 40012.html

best site
http://www.flowers-shop.org/

good site
http://www.pokkers.org/

nice site
http://www.pregnancy.net.in/

Thanks for the cogent summary- this entry was the first hit on a google search for Association and Linkage Studies, and was exactly what I was looking for. Very helpful.