Nested random effect in ancom-bc for differential abundance analysis

Hello Everyone,
I would like to ask a question about to nest random effect in the model for ancom2 analysis. To start , my involves four different strains of Daphnia magna expsode to a single concentration of a test chemical across 3 time points as shown in the excel sheet attached. I have done a PERMANOVA (Bray Curtis) on the data set using the model:
test.adonis <- adonis (dist ~ Treatment * Day * Genotype + (1|Genotype/Replicates), data = metadata, strata = metadata$Replicates)
where Replicates is nested within Genotypes.
I would like to do the same for differential abundanace analysis with ancom-bc having a a nested random effect in the model.
I have created this model:
output <- ancombc(phyloseq = ps.taxa.sub, formula = "Genotype+Day+Treatment",
p_adj_method = "holm",
zero_cut = 0.90,
lib_cut = 1000,
group = "Genotype",
struc_zero = TRUE,
neg_lb = TRUE,
tol =1e-5,
max_iter = 100,
conserve = TRUE,
alpha = 0.05,
global = TRUE)
But I don't know how include the nested random effect in the model. I would appreciate all your feedback on this.Thank you

1 Like

For future users, ANCOM-BC does not support this yet

nested random effects, which the current version of ANCOM does not support it, unfortunately

GitHub issue x-ref