Correlations in bivariate models and other things
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Correlations in bivariate models and other things



    As my graduate Thesys I'm evaluating a provenance-progeny test of
Eucalyptus globulus. I'm using the animal model (individual tree model)
to obtain the genetic values for every tree in the trial (actually two
trials).

    The command file is the following:

Prueba de evaluacion de ensayos de E globulus de Mel-Mel
 arbol !P                    # tree
 familia 199              # family (i.e. dam, cause thee families are
half-sib families)
 localidad 2               # site
 bloque 5                   # blocks
 procedencia 26        # provenance
 altura                       # height
 dap                          # diameter
pedigree.txt !alpha !make
datosmm.txt !skip 1
altura dap ~ mu !r arbol bloque procedencia bloque.procedencia
bloque.familia
!ASUV    # without this specifications the I get prolems

    Although I want to calculate the heritability for every variable
(height and diameter) and the genetic (additive) correlations between
height and diameter.

    The .pin file is as follows:

F varfen 2 + 3 + 4 + 5 + 6
F vargen 1 * 0.35
H h2 8 7

    My questions are:

1. How can I calculate two heritabilities (one for each trait) ?

2. How can I calculate the correlation between height and diameter using
"R label a ab b" ?

3. In Eucalyptus sp., there is about 20% selfing (with a relationship
ratio of 0.35). So for calculating the additive variance (in half-sib
families) you have to multiply the family variance by 2,86 and not by 4.
But in the animal model model you get the additive variance direct from
the model. However it must be correct, having in mind the 20% selfing.
How can I do that ?

    I'll be gratefull if you can help me.

Hermann Balde    cbalden@entelchile.net
Facultad de Ciencias Forestales
Universidad de Chile
Santiago - Chile







--
Asreml mailinglist archive: http://www.chiswick.anprod.csiro.au/lists/asreml