Cannot cache function 'rdist': no locator available for file...

Technical Support
1

Hi, I'm new to QIIME2 and need to use it to analyse some 16S amplicon sequencing data.

I am trying to run the following script to convert my fastq files to .qza:

generate qza file

qiime tools import
--type 'EMPPairedEndSequences'
--input-path /files/user/qiime2_bacteria/sequences/BQ1F15_TEST
--output-path /files/user/qiime2_bacteria/sequences/BQ1F15_TEST/demux.qza

--source-format CasavaOneEightSingleLanePerSampleDirFmt

output: demux.qza

I've tried a few different things, including switching the type.

However, I'm getting the following error:

"An unexpected error has occurred:
cannot cache function 'rdist': no locator available for file '/home/user/miniconda3/envs/qiime2-amplicon-2024.5/lib/python3.9/site-packages/umap/layouts.py'

See above for debug info."

My QIIME 2 is installed through 'conda' and is version 2024.5. Running through MobaXTerm on a hpc.

It could be some simple scripting mistake, but as someone new to the software I've tried a few different approaches from different tutorials and this error seems to be the consistent issue for me.

Thanks

1 Like
2

Hello @Joeee,

In what sort of environment are you running this command? In a docker container, on your laptop, something else?

3 Likes
3

Hi, I installed Qiime2 myself on a server, and the scripts are being ran through that.

I however realised there are pre-installed software on the server too (Qiime2-version 2023.7) but it comes up with the same error regardless. It may be an issue with my code so I'll keep trying different methods to see if something solves it.

4

Hello @Joeee,

Did you install qiime2 using a docker container or is it a native installation? This issue has arrived in the past when people have run out of disk space, whether in a docker container or on a shared compute resource, such as a university cluster. Could it be a disk space problem?

2 Likes
5

Segmentation fault (core dumped) - Technical Support - QIIME 2 Forum

Not sure how relevant this is, but as I mentioned earlier I sometimes get a 'Segmentation Fault'. If the Segmentation fault is occurring due to RAM issues (I should have enough, maybe the allocation is different than I'd expect) then could that be associated with the error?

6

Thanks for the response, it's on a university cluster though the set of directories I'm using should allow for over 500gb of storage (these total files are somewhere around 70gb). I have a separate install on my personal laptop (using WSL) which I can try as well.

I'm not entirely sure, but I think it's not ran through a docker. I have to load in the modules myself which are pre-installed on the cluster but tutorials for the module command make no mention of dockers.

module load bioinformatics
module load qiime2/2023.7

qiime tools import
--type EMPPairedEndSequences
--input-path raw
--output-path raw.qza

This script should in theory import the modules then process the raw data to .qza files. I noticed that when I try run the script remotely, I end up getting a Segmentation fault that I need to restart the session to fix.

7

I'm on HPC with a similar setup. They use conda install natively, then wrap it in an envmodule so users can module load qiime2/yyyy.m

So while I can't be sure, there's probably no Docker. :no_entry_sign: :whale:

Thank you for keeping us updated while the team investigates this!

1 Like
8

I've tried running the code through an alternative method (using a manifest.txt) and it seems to have worked.

I ran it directly as commands rather than in a script and on a different directory. It's probably irrelevant in terms of storage as the new directory has far less space than the previous one (60gb versus 500gb).

Here's the steps that worked for me in case its of any relevance:

  1. Module load bioinformatics

  2. Module load qiime2/2023.7

  3. qiime tools import --type 'SampleData[PairedEndSequencesWithQuality]' --input-path manifest.txt --output-path paired-end-demux.qza --input-format PairedEndFastqManifestPhred33V2

1 Like
9

I'm glad you got it working!

Would you like us to keep this issue open and investigate that rdist error?

This may be a seperate issue, but while we are here, we just updated Qiime2 with lots of new features (and some bug fixes). :bug:

Asking your HPC team to provide a version may be helpful to you and your collegues!

10

I think since I've circumvented the issue it shouldn't be a priority, but it'd be nice to understand the reasoning behind the error if possible. However as it's more for my personal understanding, I'm completely fine if you want to close the issue! It might just be on my end rather than Qiime-specific.

We've requested an updated version of Qiime2 today!

Thank you again for the help!

1 Like
11

okay! I'll unqueue for now

(My thinking here is your specific issue has been resolved, and we can always reopen / rediscover this thread the next time the issues emerges)

1 Like
12

This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.