asreml-r bivariate glmm

Hi,

I am working on the QG of life history trait corelations, and am planning to combine glmm animal model analyses with simulation-based sensitivity analyses. My first analyses are just tests to learn how to code analyses for asreml-r.

I have tried to fit a model of two poisson traits traits to some simulated data (code below). Once again I've made this work for univariate analyses, but am stuck for the bivariate analysis. Any help on this particular error will be greatly appreciated. Additionally, how would I (and can I) fit glmm models where different traits come from different distributions? Also, is there anywhere I can look for more asreml-r examples beyond what the manual provides?

Thanks,

Michael

Code:
> pedigree<-as.data.frame(read.table(file="pedForAsr.txt",header=TRUE))
> data1<-as.data.frame(read.table(file="testphen.dat",header=TRUE))
>
> P1<-as.integer(exp(data1$PHEN1/2))
> P2<-as.integer(exp(data1$PHEN2/4+1.2))
> table(P1)
P1
  0 1 2 3 4 5 6 7 8 9 10 11 13 17
223 108 44 26 8 5 7 7 1 2 3 1 4 1
> table(P2)
P2
 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 18 20 21 22 23
10 71 90 77 67 40 20 13 13 8 9 6 2 4 2 2 1 1 1 1 1 1
> data<-as.data.frame(cbind(P1,P2,pedigree[,]))
> data$ID<-as.factor(data$ID)
>
> test.ainv<-asreml.Ainverse(pedigree)$ginv
>
> test.run<-asreml(cbind(P1,P2)~trait,
+ random=~us(trait,init=c(0.1,0,0.1)):ped(ID)+us(trait, init=c(2,0,2)):ID,
+ ginverse=list(ID=test.ainv),
+ rcov=~units:us(trait, init=c(2,0,2)),
+ family=asreml.poisson(),
+ data=data)

asreml(): 2.00 Library: 2.00bi Run: Mon Mar 23 17:55:41 2009
Equations: 1762 (2 dense)
Fault 1 Data structure does not suit GLMM
Finished on: Mon Mar 23 17:55:41 2009
 
Error: Abnormal termination
Data structure does not suit GLMM

-------------------- m2f --------------------

Sent using Mail2Forum (http://www.mail2forum.com).

Read this topic online here:
http://www.vsni.co.uk/forum/viewtopic.php?p=429#429

-------------------- m2f --------------------
Received on Tue Mar 23 2009 - 18:07:39 EST