Entering edit mode
10.0 years ago
rmrf36
•
0
Hi all;
Problem : Assume a diploid locus contains 2 alleles which control a trait A(i,j). Assume that the allele I has a normal distribution in one parent. but this allele has binomial distribution in other parent. after complete crossing of parents, we have F1 generation. What is the probability distribution of F1 ? Is it predictable considering the distributions of parents? Can we estimate the probability distribution for F1 to accomplish significance tests?
Regards
What would it even mean for an allele to have a normal versus binomial distribution in a parent? Are the parents some sort of complex chimeric organisms?
BTW, you can always take the "I'm not a statistician" route and perform a Monte Carlo simulation. You'll get approximately the same answer in either case, with the difference dependent on the number of simulation iterations.
I am not going into the validity of assuming different distributions for the two alleles on the same locus, but it sounds like you are looking for a bayesian hierarchical model (which can be simulated using a Markov Chain Monte Carlo, as Devon Ryan mentioned). That is, if you can come up with a likelihood function for your offspring.