Hi All,
I'm performing admixture analysis using the BMIX method and I need to estimate the effective number of tests. Some background, I have estimated the genotype of loci on chromosomes for a set of hybrid individuals, where 0 = homozygous locus from species A, 1 = heterozygous locus (spA, spB), and 2 = homozygous locus for species B. The genotypes of the loci (in terms of their ancestry) is strongly autocorrelated and I need to input the amount of autocorrelation across my set of inds into a model.
The method I'm using is very detailed on how to set up every aspect of the model except for estimation of the effective number of tests. I've hunted down the methods I think they used but I cannot find a connection from the output of the effectiveSize() function in coda to the effective number of test that I think are reasonable.
I'm conditioning my exceptions upon the number of crossing over events on a chromosome for a specific individual. For example, in the genotypes below, there are 5 blocks of autocorrelated ancestry, and 5 independent pieces of chromosome. In my mind, the effective number of tests = 5.
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
This string of 0's & 1's represents a set of SNPs along a chromosome.
According to effectiveSize(), there are 4.658942 tests here. That's pretty close to five, but I'm still confused. I have an individual with 2 blocks of ancestry and about 48 tests according to coda's effectiveSize().
Can anyone with experience shed some light on how to interpret the output of effectiveSize, estimating autocorrelation and the effective number of tests from genomic data?
Thanks!
