I have been using the parallel method in the pin file for calculating
correlations etc from multi-trait analyses, and this works great for series
of traits which have the same random effects models. However, it is far
from simple where there are several traits which have different models for
random effects. Is there a shortcut method in this case for the pin file???
Or, should you set up the analysis to give you the same models for each
trait, but fixing some to very low values for specific cases (I imagine
this might cause problems with convergence etc?).
For example: traits 1 2 3 4 and 5 fitting additive for all, but c2 only for
traits 1 and 2. With identical random effects models the phenotypic
variance would be simply 1:15 + 16:30. Now that we have c2 for two traits
only I find I have to build more than one phenotypic covariance matrix, and
remember which one is correct for which traits.....Is there something I
have missed in the manual which allows you to add up the matrices correctly?
Hope you can help with simplifying this....
Kim Bunter (M.Rur.Sc)
Animal Genetics and Breeding Unit
University of New England
Armidale, NSW, 2351
Ph (ISD): -61-2-67733788
Fax (ISD): -61-2-67733266
Asreml mailinglist archive: http://www.chiswick.anprod.csiro.au/lists/asreml