repeated and/or nested measures dbRDA/PERMANOVA

[hugoeira]

Hi all,

I hope this post is relevant for the discussions here.

I am working on the microbiome of birds(nestlings). At this moment I am trying to look at how beta diversity (bray-curtis and unifrac) varies with age and between families (nests)

My dataset is comprised of individuals sampled at two time points (t1 and t2). Age of each individual varies at these time points (its a wild population age can only be estimated after capture). In sumary: each individual is sampled twice.
But some individuals are from the same nest, some are from diferentt nests.

In total: 112 individuals (224 sampling points) belonging to 55 nests.

I was trying to approach this with PERMANOVA but as I was reading some posts here it seems that dbRDA might be a better option.
PRIMER seems to be bale to deal well with random effects and nested designs but its very expensive.

My first problem is how to deal with repeated measures. The solution seems to be blocking for individual ID:

ctrl <- how(nperm = 1000)
setBlocks(ctrl) <- with(metadata, individual_ID)

Second problem is how to account for the effect of nest and the fact that some birds belong to the same nest. I have been playing with this:

ctrl <- how(within = Within(type = "free"), plots = Plots(strata = metadata$nest, type = "none"))

but here i am not sure what the permutation restriction is doing.

So my final model would be:

db_rda <- dbrda(asv_table ~ sampling_point + nest + sex + year + Condition(individual_ID) , data = metadata,
method = "bray", add = "lingoes")

Should nest be included as condition as well? is ther a way to tell the model that individual_ID is nested within nests?

Last question should I include sampling point or age (in days)? Does dbRDA deals well with continuous variables?

Really sorry for the long post and all the questions. I have been trying to solve this problems for about 2 weeks now and is driving me crazy.

Kind regards,
Hugo