LOD score permutation too high?

[Tiago Ribeiro]

Hi,

I am running permutations for scantwo, I have about 250 RILs, and chromosomes of around 500 cM. My LODfull score after 2000 permutations was around 14, I looked at the LOD distribution and saw that it was not unimodal, so I removed outlier individuals and re-ran everything. Now I am getting LODfull of 10.5 (at 5%) and 10 (at 10%), the distribution of LODfull scores now is unimodal. Does it seem too high? Is there any intuition of how high LODfull permutations should be?

--

Tiago da Silva Ribeiro

Ph.D. Candidate

Pool Lab

Department of Integrative Biology, UW-Madison

Laboratory of Genetics, UW-Madison

[Karl Broman]

Significance thresholds for two-dimensional scan can be large. They depend principally on genome size (which is greatly expanded in RILs) and then on type of cross, analysis method, and density of markers. Also some influence from pattern of missing genotype data, the phenotype distribution, and sample size.

karl