David,
Neil is correct. -genhw- works for multiallelic loci.
For multiallelic loci it performs an asymptotic test so it may not perform well with sparse data. See Cleves M., Hardy-Weinberg equilibrium test & allelle frequency estimation STB-48 sg110,3/99 (STB Reprints Vol 8, pages 280--284) for more details.
Make sure that you install the latest copy of the program from my website "www.biostat-resources.com/stata".
Regards,
Mario.
-----------------------------------------
Mario A. Cleves, Ph.D.
Associate Professor and Senior Biostatistician
Department of Pediatrics
UAMS College of Medicine
Arkansas Center for Birth Defects Research and Prevention
11219 Financial Centre Parkway, Suite 250
Little Rock, AR� 72211
Telephone: 501-364-5001
Fax: 501-364-5107
Email: [email protected]
-----------------------------------------
-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Neil Shephard
Sent: Saturday, March 12, 2005 6:48 AM
To: [email protected]; [email protected]
Subject: Re: st: Hardy-Weinberg for multiallelic loci
On Fri, 11 Mar 2005 15:54:39 -0800, Doody, David R <[email protected]> wrote:
>
> I searched the archives for an answer to this question, but was
> unsuccessful. Sorry if it's been discussed before. I'm trying to test my
> case-control data to confirm they're in Hardy-Weinberg equilibrium. I can
> use genhw for SNPs, but I have a few multiallelic loci. I've seen various
> other packages which seem to use a method proposed by Guo and Thompson
> (using Markov chain methods?), but they require extra effort to get the data
> into a format for those packages to work. I'm wondering if anyone has
> developed programs for this method within Stata? Or is there another method
> which is preferred for testing HWE for multiallelic loci that I should know
> about? Thanks in advance.
>
The -genhw- command will work with multi-allelic loci (see the help
page for details of format). You don't explicitly state the format
your data is in, but I suspect it may already fit the required format
(i.e. two variables for each locus, with the alleles an individual
carries listed in each variable).
The reported statistic is simply the Pearson Chi^2 for the observed
and expected genotyped frequencies.
HTH's
Neil
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
