HCGB-IGTP/XICRA

miRNA module not finding joined reads

Closed this issue · 2 comments

jonahcullen commented

Hello, I am trying to use the miRNA module following prep, trim, and join (starting with SE reads) as

XICRA miRNA \
    --input Prep/ \
    --output_folder Prep/ \
    --species eca \
    --database miRBase \
    --software optimir \
    --single_end

everything up to this point worked as expected (NOTE - fastq-join had to be installed separately and was not included following the conda-based installation). Upon executing the above code, I get the following error for each of the included test samples (n=2)

** ERROR: Only 1 fastq file is allowed please joined reads before...

When I run with --debug I can see that the table being built are indeed pulling both R1/R2 instead of the trimmed and joined FASTQs which exist as expected in Prep/data/SAMPLE/join/SAMPLE_trim_joined.fastq.

I'm sure I'm just missing something but please let me know if there are any other logs you would like to see.

JFsanchezherrero commented

Hi Jonah,
Thanks for pointing out about the fastq-join missing dependency in the conda environment. I also found featureCounts was missing in a completely fresh installation so be aware of that too. I have recently updated both the conda environment file and the python code, so you might find useful to re-install (at least the python code, make sure version >v.1.4.5 by typing pip install XICRA)

Regarding the question you mentioned and the error you encountered using your data I suggest to remove some specs from you command: --output and --single-end

I guess you should have generated a project folder containing the reads, qc results, and trimmed reads right? So you only require to use --input flag as output results would be generated too there. Regarding the --single-end flag, if you have already joined reads, you should not specify this flag anymore.

Your command would be

XICRA miRNA \
    --input Prep/ \
    --species eca \
    --database miRBase \
    --software optimir \

I have to say I haven't tested in any other species than human but it should work fine. Please let me know if it not works proprely.

Also, I prefer to use miraligner than optimir, although both are implemented. We found better performance using simulated reads in the XICRA paper (https://doi.org/10.1186/s12859-021-04128-1)

I have recently updated the code and I have added an example. Please run XICRA test to get a set of reads to test the pipeline. Read the commands included in the test_subset.sh generated and find the appropiate steps to reproduce the results.

Please, do not hesitate to contact me for further details.

Best regards,
Jose

jonahcullen commented

Removing --single-end worked, I must have misunderstood the help prompt. Thank you again!

  • 👍1
  • ❤️1