[R-sig-ME] ziPoisson heritability MCMCglmm

[Mazé E]

Hi,
I am currently trying to fit a quasi-poisson distribution with my data (counts
variable, which contains 50% of zero) with MCMCglmm so as to estimate the
heritability of a morphological trait. Previously I used a "Poisson"
distribution, but to be more exact I have to use a ziP distribution. However, I
am not comfortable with bayesian priors and  the syntax of the models in
MCMCglmm, and I don't understand all the terms of the outputs
("summary.MCMCglmm(object)").
Can someone help me to 
1)asses if the prior used is non-informative?
2)if the ziP model syntax have the same signification as the Poisson model?
3)explain which terms I have to use so as to calculate the heritability?

My codes are

#Poisson model

prior1.7 <- list(G = list(G1 = list(V = 1, n = 1.002), G2 = list(V = 1,n =
1.002), G3 = list(V = 1, n = 1.002),G4 = list(V = 1, n = 1.002),
 G5 = list(V = 1, n = 1.002)), R = list(V = 1, n = 1.002))

modelDEG <- MCMCglmm(DEG ~ PAR +SIZE, random = ~ animal + ANNEE + STATION +
ANNEE:STATION , pedigree = Ped, data = Data, nitt = 1000000, thin = 1000, burnin
=  50000,prior = prior1.7, verbose = FALSE,family="poisson")

Heritability=mean(modelDEG$VCV[, "animal"] / (modelDEG$VCV[, "animal"] +
modelDEG$VCV[, "units"] + modelDEG$VCV[, "STATION"]+ modelDEG$DVCV[, "ANNEE"]+
modelDEG$VCV[, "ANNEE:STATION"]))

#ziPoisson model

zi.prior <-  list(R = list(V = diag(2), n = 1.002, fix = 2),
 G = list(G1 = list(V = 1, n = 0.002),
 G2 = list(V = 1, n = 0.002),
 G3 = list(V = 1, n = 0.002),
G4=list(V=1,n=0.002)))

model1 <- MCMCglmm(DEG ~ trait+trait: PAR +trait:SIZE, 
random = ~ animal + ANNEE + STATION + ANNEE:STATION,
rcov=~us(trait):units,
pedigree = Ped, data = Data, 
prior=zi.prior,
nitt = 1000000, thin = 1000, burnin =  50000,
verbose = FALSE,family="zipoisson")

Heritability= ????


Many Thanks in advance