Date:         Fri, 31 Aug 2007 11:40:47 -0400
Reply-To:     Kevin Roland Viel <kviel@EMORY.EDU>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Kevin Roland Viel <kviel@EMORY.EDU>
Subject:      Re: AIC versus c statistic in Proc Logistic
In-Reply-To:  <011401c7ebdc$9cbec930$c12fa8c0@HP82083701405>
Content-Type: TEXT/PLAIN; charset=US-ASCII

On Fri, 31 Aug 2007, Mary wrote:

> It is good to see that you are in Genetics.  My "subjects" are
> independent- they are donor eyes of people who have died that had a form
> of blindness (AMD), along with "controls" of donor eyes that did not
> have blindness; and more definition in which genes/snps discriminate
> between those who get the disease versus those who don't will hopefully
> help family members determine whether they are at risk or not (but I
> don't have family members in this data set, thus I have independent
> observations).
>
> I got the book from Amazon; I went to buy the SAS/Genetics 9.1 Users's Guide and Amazon recommends it as a companion book.


One eye per subject?  This is the so-called association study.  Its a big
trend at the moment.  One interesting study is:

Complement factor H polymorphism in age-related macular degeneration.
Klein et al. Science. 2005 Apr 15;308(5720):385-9.

I find it interesting because out of 116,204 SNPs, only two had low
p-values.  See their figure to get a full feeling for this.

Amos has just published a paper discussing GWAS (genome-wide association
studies) that may be of interest.

Kevin



Kevin Viel, PhD
Post-doctoral fellow
Department of Genetics
Southwest Foundation for Biomedical Research
San Antonio, TX 78227