Hi all !
I'm a bit lost with the genetics! Please help!
We have a population of about 800 animals (three different breeds) on which 2 indices of performance where mesured. Selection of a certain number of non-related animals (different parents) was made on the basis of *extreme* results on both indices (superior : quadrant ++ or inferior : quadrant --), favoring the inferior when two animals with a paternal or maternal link were found extreme, considering the rarity of the inferior trait. Unfortunatly, the bivariate distribution turned up to be circular (non ellipsoidal) and extreme cases then didn*t seem to be so extreme after all (which raises many other questions).
Nevertheless, the genotype of each selected animal was then obtained and we looked at the allelic frequencies per race and sex. The results show very small frequencies of the mutant allele. We are also interested in calculating the average of each of the three groups (nn, nN, NN) on both indices that were use in the prior selection of the animals (the purpose of that not being very clear to me).
The question I*m facing is : What kind of inference am I in title to apply for the frequencies obtained regarding the sampling that was done, e.g. chosing extreme cases to amplify possible genomic differences. And how can I interpret or even test for significant differences in the calculated means?
Any help will be much appreciated.
Steve Méthot
Statisticien
Agriculture et agroalimentaire Canada
C.P. 90, 200, Route 108 Est
Lennoxville (Québec) J1M 1Z3
téléphone: (819) 565-9174 ext. 262
télécopie : (819) 564-5507
courriel : [log in to unmask]
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|