SVARET FRÅN GEORGE STRAIN

RE: Confounded‏

 George M Strain

Till Bodil Carlsson
I’m not a statistician myself. However, I think that part of the problem is that they are trying to quantify the contributions of various phenotype components to deafness, but I think that logistic regression models are based on an assumption that the various factors being included in the model are independent; so-called interaction terms are meant, I believe, to show how much they are inter-related. The items they modeled are highly inter-related, or likely caused by the same base genetic factor(s). I don’t think they can say, if indeed they do so, that a white head is more of a risk for deafness than the presence of the merle gene. The white head could be due to merle or recessive S alleles or both. We don’t know.
 
I agree that there is no way that they can justify the statement that “S locus is not thought to be involved in the regulation of deafness in Border Collies”. I would argue that they are absolutely wrong on that. Otherwise, how do we get deafness in BC that don’t carry merle.
 
________________________________________________
George M. Strain
Professor and Interim Head
Comparative Biomedical Sciences
Louisiana State University School of Veterinary Medicine
Baton Rouge, LA 70803
Voice 225-578-9758 Fax 225-578-9895