Killed qiime feature-classifier fit-classifier-naive-bayes

Hello, cordial greeting

I am new to qiime and I am doing a bacterial microbiome analysis of pigeon fecal matter. The sequences were initially treated with the FLASH program. These sequences were paired-end, but, after FLASH treatment they were treated as single-end. I am using Linux as a Windows subsystem Ubuntu 22.04 with WSL2 and the version of Qiime2 is qiime2 2024.2
The problem I have had is that when executing the command:

qiime feature-classifier fit-classifier-naive-bayes
--i-reference-reads ref-seqs-300-500.qza
--i-reference-taxonomy silva-138-99-tax.qza
--o-classifier classifier-300-500.qza

The error is generated: Killed
I have been reading about the subject and I have found that this happens because the command to execute requires good RAM. Therefore, I checked the PC resources and it has 8GB of RAM of which WSL2 has 6GB for use.

I have two concerns, the first is how much RAM does this command need? and second, how can I assign more GB of RAM to WSL2?

Hello @LisethBolCol,

It's difficult to say how much RAM this command will need because it depends on a number of factors. As far as allocating more RAM to WSL2, you should read the WSL2 documentation to find out. With only 8GB total on your computer it might unfortunately just be the case that even with all memory allocated to WSL (or as much as possible) you won't able to run the command with these inputs.

1 Like

Hi
I have already looked at the page that shows how to install WSL, but there is no concrete explanation of what commands to use to make a new RAM allocation. I have read in some bioinformatics forums and the commands do not work for me.

Hi @LisethBolCol,

Check out the WSL docs on their available settings - I haven't used WSL myself, but it seems like the swap parameter might be what you want to modify to adjust your RAM allocation.

Cheers :lizard:

Thank you very much, I managed to run the command on a desktop computer with 16GB of RAM and I observed that this command uses a little more than 12GB of RAM

1 Like