AMCOM-BC vs ANCOM-BC2 and number of significant results

[mwhite942]

I have looked, but haven't been able to find very much information about the differences between ancombc() and ancombc2(). Could someone provide more information about what has changed between these two functions, or point me to a resource that details the differences?

I'm especially wondering because I had used ancombc() on my dataset recently, but was thinking I should use ancombc2(), as it is the updated version. However, when I ran a test on the exact same data for the same variable, I had 1869 significant OTUs when using ancombc2(), but only 46 when using ancombc(). I am determining significant OTUs based on the q value for habit (my variable of interest). (For reference, I have 12,412 OTUs in the analysis after the prevalence filtering.) I am running these analyses in RStudio, with the following code: habit.out4 <- ancombc(data = v4v5, tax_level = "zotus", formula = "habit", p_adj_method = "holm", prv_cut = 0.10, lib_cut = 0, group = "habit", struc_zero = TRUE, neg_lb = FALSE, tol = 1e-5, max_iter = 100, conserve = TRUE, alpha = 0.05, global = TRUE, n_cl = 1, verbose = TRUE) for ancombc() and habit.out3 <- ancombc2(data = v4v5, fix_formula = "habit", rand_formula = NULL, p_adj_method = "holm", pseudo_sens = TRUE, prv_cut = 0.10, lib_cut = 0, s0_perc = 0.05, group = "habit", struc_zero = FALSE, neg_lb = FALSE, alpha = 0.05, n_cl = 2, verbose = TRUE, global = FALSE, pairwise = TRUE, dunnet = FALSE, trend = FALSE, iter_control = list(tol = 1e-5, max_iter = 20, verbose = FALSE), em_control = list(tol = 1e-5, max_iter = 100), lme_control = NULL, mdfdr_control = list(fwer_ctrl_method = "holm", B = 100), trend_control = NULL) for ancombc2().

It seems odd that I would get such vastly different numbers of significant OTUs in the two analyses. (38 of the OTUs are shared, but 8 of the 46 from ancombc() are not included in the results from ancombc2().) I am wondering if I am not using the correct parameters for one of the functions, or if something else is going on. 1869 significant OTUs seems too high, additionally. Any ideas as to what is going on?

Thanks in advance.