Reproducing results with ancombc()

[ellimakar]

Hello :)

I started exploring the ANCOM-BC and I am trying to reproduce the results from the article Analysis of compositions of microbiomes with bias correction when comparing MA vs US at the 0-2 age group by using the ancombc() function. However, I get different results than those presented in the article...Not sure what I am missing but the code I am using is the following (after creating the phyloseq object):

test <- phyloseq(obs.abn_phyl, meta.data_phyl)

ancombc(phyloseq = test, formula = "country",
p_adj_method = "bonferroni", zero_cut = 0.9, lib_cut = 1000,
group = "country", struc_zero = TRUE, neg_lb = TRUE, tol = 1e-5,
max_iter = 100, conserve = FALSE, alpha = 0.05, global = TRUE)

Thank you in advance!!
Elli

[FrederickHuangLin]

Hi Elli,

Thanks for your interest in ANCOMBC!

The results from the paper you mentioned were produced using the standalone codes (ANCOMBC R package was not available at that time), which can be fully reproduced by this repo.

We have updated the ancombc function since publication. Thus, it is expected to see some discrepancies, but the difference is supposed to be minimal.

Best,
Huang