"Killed 9" Error (When Running Adonis Command)

[Juston]

Hello,

I am trying to run the following command, but keep getting a "Killed 9" Error:

qiime diversity adonis
--i-distance-matrix /Users/Jaco/Downloads/core-metrics-results/unweighted_unifrac_distance_matrix.qza
--m-metadata-file /Users/Jaco/Downloads/metadata_with_oxalatebins_clean_caox_indices.txt
--p-formula Oxalic_acid_to_calcium_ratio+oxalic_acid_plus_calcium+oxalic_acid_in_mg__vioscreen_foodcomponents
--p-permutations 999
--p-n-jobs 1
--o-visualization /Users/Jaco/Downloads/adonis_oxalate_unwUF.qzv

It appears as though there is a bug in the adonis command, since it is consuming way too much memory. I have 16GB of memory on my laptop and have about 2,000 samples in the distance matrix? My colleague just ran the command on his computer, it also ran out of memory w/ 64GB!

Will try again on 250GB to see if that makes a difference, but curious to know if this came up for anyone else?

[jwdebelius (Justine Debelius)]

Hi @Juston,

Welcome to the forum!

Its been a while since I ran a distance matrix of a thousand samples; I remember a long run time but not a memory error.

I would also make absolutely sure that you're passing what look like continuous values as continuous values by specify in your metadata. You might be running into issues because you're trying to model something with 1500 or 2000 degrees of freedom.

You might also consider co-linearity. It looks like your variables are similar. You might look at whether they're highly correlated and try to select a representative variable rather pushing all three in.

Best,
Justine