NULL results for global, pairwise and pattern analysis

[otaviolovison]

Hello!

I am running alongitudinal analysis with ANCOMBC2 using the code below, and the global, pairwise and pattern are NULL in the output:

set.seed(123)

genus_output = ancombc2(data = tse, assay_name = "counts", tax_level = "Genus", fix_formula = "Age + Sex + Exam + sample.type", rand_formula = "(Exam | patientID)", p_adj_method = "holm", pseudo_sens = TRUE, prv_cut = 0.10, lib_cut = 1000, s0_perc = 0.05, group = "sample.type", struc_zero = TRUE, neg_lb = TRUE, alpha = 0.05, n_cl = 20, verbose = TRUE, global = TRUE, pairwise = TRUE, dunnet = TRUE, trend = TRUE, iter_control = list(tol = 1e-2, max_iter = 20, verbose = TRUE), em_control = list(tol = 1e-5, max_iter = 100), lme_control = lme4::lmerControl(), mdfdr_control = list(fwer_ctrl_method = "holm", B = 100), trend_control = list(contrast = list(matrix(c(1, 0, -1, 1), nrow = 2, byrow = TRUE), matrix(c(-1, 0, 1, -1), nrow = 2, byrow = TRUE), matrix(c(1, 0, 1, -1), nrow = 2, byrow = TRUE)), node = list(2, 2, 1), solver = "ECOS", B = 100))

I believe that the cause of this is the random_formula, where I am 'blocking' the temporal variable 'Exam' and the patient ID. In addition, I have another 'issue' in this study design, where the same patientID have both group (sample.type) categories represented, since the samples where collected from the same patient but from different sites (health and disease), so, we have a 'double' autocorrelation problem. I tried to remove 'Exam' from the random_formula, but the global, pairwise and pattern are still NULL, which I believe that is being caused by this 'double' autocorrelation.

I would like to discuss if:

• Is my modeling correct?
• Should I change the categories of the variable patientID to differentiate between healthy and diseased samples (due to blocking), even though I already have the group variable (sample.type) showing this difference to the model?

Thanks in advance!