q-values (adjusted p-values) too low for continuous variable

[Pantalius]

Hi @FrederickHuangLin,

I used ANCOMBC2 to test the association between microbiome and a continuous outcome (bone mineral content) using data from phyloseq object and my adjusted p-values are too low (e.g. p=3.79e-302 which looks too good to be true). p-adjustment was by "holm". I tried the other adjustment arguments methods like "bonferroni" etc and had similar results. In addition to my outcome variable "tblh_aBMC_6yr", I included other categorical and continuous covariates.

Here is the code:
ancom_4y_aBMC <- ancombc2(data = studypop_phy_4y, assay_name = "counts", tax_level = "genus",
fix_formula = "tblh_aBMC_6yr+sex+race+SES+height_6yr.x+age_6yr+tblh_BFM", rand_formula = NULL,
p_adj_method = "holm", pseudo_sens = TRUE,
prv_cut = 0.10, lib_cut = 1000, s0_perc = 0.05,
group = NULL, struc_zero = FALSE, neg_lb = FALSE,
alpha = 0.05, n_cl = 1, verbose = TRUE,
global = FALSE, pairwise = FALSE, dunnet = FALSE, trend = FALSE,
iter_control = list(tol = 1e-2, max_iter = 20,
verbose = TRUE),
em_control = list(tol = 1e-5, max_iter = 100),
lme_control = NULL,
mdfdr_control = list(fwer_ctrl_method = "holm", B = 100))

1. Is there something wrong with my arguments or I'm I missing something else?
2. Your recent paper on ANCOMBC2 (Lin & Peddada, 2024) largely addressed categorical variables (multigroup analysis). Would you recommend using this method where my primary outcome is a continuous variable?
3. Also, just out of curiosity, I inserted "relative_abundance" argument in the assay_name and the output was exactly the same as when I used "counts". I know that it is not logical to use relative abundance since we are dealing with complete cases only (excluding zero counts). If this is the case, should the model accept "relative_abundance" as a valid argument?

I would appreciate your feedback on these concerns.

Thanks
Pantalius