random effect of clade for species in a phylo tree

[Jigyasa3]

Hey!

I am following the simulated examples from https://leanpub.com/correlateddata/read#leanpub-auto-trait-by-environment-interactions. I was wondering if it is possible to examine the random effects of clade of each species (as a factor).

I tried with the simulated example in section 4.6 of the above link, but I keep getting the following error-

#simulated data- #Simulate the data- nspp <- 20

Simulate a phylogenetic tree

phy <- compute.brlen(rtree(n = nspp), method = "Grafen", power = 1)
sd.env <- 1
env1 <- rnorm(nspp, sd=sd.env)
env2 <- rnorm(nspp, sd=sd.env)
lineage<-rep(c("s1","s2","s3","s4"),times=5)
sd.trait <- 1
trait1 <- rTraitCont(phy, model = "BM", sigma = sd.trait)
trait2 <- rTraitCont(phy, model = "BM", sigma = sd.trait)

d2<-data.frame(sp=phy$tip.label,trait1=trait1,env1=env1)
d2$Y <- b0 + b1*d2$trait1 + b2*trait2 + c1*d2$env1 + c2*env2 + f11*d2$trait1*d2$env1 + f21*trait2*d2$env1
d2$lineage<-as.factor(lineage)

d2%>%head() sp trait1 env1 Y lineage t7 t7 0.1155387 0.3428021 0.41741713 s1 t4 t4 -0.0228887 -0.2150611 0.05225095 s2 t8 t8 -0.4770557 0.5494478 0.43393900 s3 t9 t9 -1.1772624 0.2776693 -0.43220921 s4 t13 t13 -0.4740025 0.4773304 -0.23581239 s1 t14 t14 -1.2359375 -0.3810483 -1.48079339 s2

#model- mod.f <- communityPGLMM(Y ~ 1 + (1|sp__)+(lineage|sp__)+ (env1|sp__), tree=phy, data=d2)

#error- Error in pglmm_gaussian_internal_cpp(par = s, X, Y, Zt, St, nested, REML, : Evaluation error: objective in x0 returns NA. In addition: There were 41 warnings (use warnings() to see them)

My aim to examine the effect of each clade of species on a trait of interest. Any suggestions?

[daijiang]

Sorry for the late response. Can you paste the whole example? Also, try mod.f <- pglmm_compare(Y ~ 1 + (1|sp__)+(lineage|sp__)+ (env1|sp__), data=d2, phy = phy). Thanks.