simulation genotype data

David Duffy davidD at orpheus.qimr.edu.au
Mon Jun 7 06:34:27 EST 1999


Chi-Hse Teng (teng_chi-hse at REMOVE.lilly.com) wrote:
: Hi, I have a question about simulating sib pair data.
: My concern is, when we keep or infer the parent's marker genotype for
: simulation, the p-value we get is from the conditional LOD distribution
: given the parent's genotype.    Is there a or some parent's genotype
: will give us a distribution different from the distribution of LOD?

Well, in the limiting case of perfectly informative markers or large
samples, it shouldn't matter.  The only thing differentiating 
samples from your population will be the informativeness of the parents
included in that sample, which would be a function of presence/absence of
HWE, proportion parents typed, marker heterozygosity.  I could imagine a
small sample with an abnormally high proportion of informative parents
could have high power to detect a true linkage (your conditional simulation),
but be unlikely to be replicated by a study of the same size (uncond sim).

David Duffy.




More information about the Gen-link mailing list