Hi everyone,

I am giving a try to asreml-R v.4 for plant and animal breeding programs.

I am basically trying to translate some asreml code from the book "Genetic Data Analysis for Plant and Animal Breeding" into asreml-R v4. I am not able to define the G matrix which contain direct genetic effect (additive variance), maternal effect and the covariance between both of them. In particular, the asreml code I am trying to translate into asreml-R v4 is:

Code example 4.5 in the book:

weant ~ year sex weanage !r str(pig dam us(2) .nrm(pig))

were year, sex and weanage are fixed effects and pig (animal ID) and dam are random effects. This code defines a residual structure like this:

V = | G 0|

|0 R|

= A*V(a) A*cov(a,m) 0

A*cov(a,m) A. V(m) 0

0 0 R

where A is the numerator relationship matrix A (obtained from the pedrigree), V(a) is the additive variance for direct genetic effects, V(m) is the additive variance for maternal genetic

effects, and cov(a,m) is the covariance between the direct and maternal genetic effects. R is de residuals variance.

Can anybody help me to translate this code into asreml-R v4 code?

Similarly, an even more complex model and structure is:

weant ~ year sex weanage !r str(pig dam us(2) .nrm(pig)), us(2 !GUUU) .nrm(pig), ide(dam)

which includes maternal environmental effect into the model (also random). The variance structure for this model is:

V = A*V(a) A*cov(a,m) 0 0

A*cov(a,m) A. V(m) 0 0

0 0 I*V(me) 0

0 0 0 R

where I is the identity matrix and V(me) is the variance of the environmental maternal effects (i.e., maternal effects that cannot be associated with the pedigree information).

I would really appreciate any hint on this problem.

Thanks!!!!

Miguel