[R] help needed for HWE.exact in library "genetics"

[Neil Shephard]

You could follow the advice given when loading the library (see the code you
posted for details) and use the enhanced genetics packages to cross-validate
the results (ideally you should get the same answer).

The results not weird though.

Your working with SNPs and having a homozygote with a frequency of zero
isn't that unlikely, it depends on the minor allele frequency (MAF).  In
this instance its the controls you appear to be concerned about, and for
this locus the MAF of the minor allele is 0.0145, so in a sample of 173
individuals you would expect to see 0.0363 individuals with the recessive
genotype (basic Mendelian genetics of 173 * 0.0145^2).  Thus at best you
might expect to see one person in a sample of that size, but its not
surprising that you've not seen any.

You may also find the following references of interest...

Guangyong Zou, Allan Donner
The Merits of Testing Hardy-Weinberg Equilibrium in the Analysis of
Unmatched Case-Control Data: A Cautionary Note
Annals of Human Genetics
Volume 70, Issue 6, Pages 923-933

(basically advocates the use of a robust test for association such as the
trend test over the two-stage design of testing for HWeqm and then testing
for association).

There is follow up discussion of the article in 

Yik Y. Teo, Andrew E. Fry, Taane G. Clark, E. S. Tai, Mark Seielstad
On the Usage of HWE for Identifying Genotyping Errors 
Annals of Human Genetics
Volume 71 Issue 5 , Pages 701 - 703

...and a rejoinder from Zou and Donner...

Guangyong Zou, Allan Donner
The Reply of Zou & Donner to Teo's Commentary on their Manuscript (p
704-704)
Annals of Human Genetics
Volume 71 Issue 5 , Pages 701 - 703

Neil

Sun, Ying [BSD] - HGD wrote:
> 
> ============
>> library( "genetics" )
> NOTE: THIS PACKAGE IS NOW OBSOLETE.
>   The R-Genetics project has developed an set of enhanced genetics
>   packages to replace 'genetics'. Please visit the project homepage
>   at http://rgenetics.org for informtion.
> 

-- 
View this message in context: http://www.nabble.com/help-needed-for-HWE.exact-in-library-%22genetics%22-tp19100974p19105112.html
Sent from the R help mailing list archive at Nabble.com.