Maaslin3 is running but is not able to generate any results

Downstream analysis and statistics MaAsLin
1

so this is my formula that im using:-

Add reads column (total counts per sample)

metadata_to_test2$reads ← rowSums(abund_matched)

fit_out <-maaslin3(
input_data = abund_matched,
input_metadata = metadata_to_test2,
output = “maaslin3_output”,
formula = ‘~ Age + Gender_code + BMI + diabetes_code + medHis_highSugar_code + antibiotics_2weeks_code + reads’,
normalization = ‘TSS’,
transform = ‘LOG’,
augment = TRUE,
standardize = TRUE,
max_significance = 0.1,
median_comparison_abundance = TRUE,
median_comparison_prevalence = FALSE,
max_pngs = 250,
cores = 1
)..and it is showing this for every amr gene family ethyltransferase (G1405)
2025-10-21 15:20:15.749003 WARNING::Fitting problem for feature 1 returning NA
2025-10-21 15:20:15.751084 INFO::Fitting model to feature number 2, 50S rRNA with mutation conferring resistance to macrolide antibiotics
2025-10-21 15:20:15.753023 WARNING::Fitting problem for feature 2 returning NA
|+ | 1 % ~01s 2025-10-21 15:20:15.754994 INFO::Fitting model to feature number 3, AAC(3)
2025-10-21 15:20:15.757686 WARNING::Fitting problem for feature 3 returning NA
2025-10-21 15:20:15.759442 INFO::Fitting model to feature number 4, AAC(6’)….and in the end I’m getting all zero values with nothing significant…where did i get wrong???

2

Hi,

Nothing looks obviously wrong at first glance. Some things to start with:

  1. What are the dimensions of your input data and how many total categories does that formula come out to?
  2. Could you send 2-3 rows/columns of the anonymized input data? There might be something funky about the data that’s not obvious from just the MaAsLin 3 call

Will