mhalushka/miRge3.0

Error when running miRge3.0: ValueError in summarize function

Closed this issue · 6 comments

shashankpritam commented

Hello! When running miRge3.0, I encountered an error during the "Summarizing and tabulating results" step. The traceback message and error indicate an issue with setting a DataFrame with multiple columns to a single column called miRNA_cbind.

The command to run the job was as follows:

#!/bin/bash
#SBATCH --job-name=miRge3_job
#SBATCH --output=miRge3_output.txt
#SBATCH --ntasks=1
#SBATCH --cpus-per-task=1
#SBATCH --time=72:00:00
#SBATCH --mem=16G

eval "$(micromamba shell hook --shell bash)"
micromamba activate mirna

miRge3.0 --samples some1.fastq,some2.fastq --mir-DB miRBase --organism-name fruitfly --outDir miRge3_output --libraries-path miRge3_Lib

But, the following error is encountered during the execution of the summarize function:

Alignment in progress ...
Alignment completed in 6255.9694 second(s)

Summarizing and tabulating results...
Traceback (most recent call last):
  File "/home/shashank/micromamba/envs/mirna/bin/miRge3.0", line 10, in <module>
    sys.exit(main())
  File "/home/shashank/.local/lib/python3.10/site-packages/mirge/__main__.py", line 166, in main
    summarize(args, workDir, ref_db, base_names, pdMapped, sampleReadCounts, trimmedReadCounts, trimmedReadCountsUnique)
  File "/home/shashank/.local/lib/python3.10/site-packages/mirge/libs/summary.py", line 744, in summarize
    subpdMapped['miRNA_cbind'] = subpdMapped[['exact miRNA', 'isomiR miRNA']].apply(lambda x: ''.join(x), axis = 1)
  File "/home/shashank/micromamba/envs/mirna/lib/python3.10/site-packages/pandas/core/frame.py", line 3940, in __setitem__
    self._set_item_frame_value(key, value)
  File "/home/shashank/micromamba/envs/mirna/lib/python3.10/site-packages/pandas/core/frame.py", line 4094, in _set_item_frame_value
    raise ValueError(
ValueError: Cannot set a DataFrame with multiple columns to the single column miRNA_cbind

Could you please let me know how to tackle this?
Thank you!

Edit: I used micromamba create mirna mirge3 to install mirge3.

arunhpatil commented

Hi @shashankpritam,

This is probably due the old version of numpy installed. Can you let me know what numpy version is installed?

Thank you,
Arun.

shashankpritam commented

Hello @arunhpatil Thank you for your prompt response.
Here's numpy info -

Name: numpy
Version: 1.23.2
Location: /home/shashank/.local/lib/python3.10/site-packages
arunhpatil commented

This version seems fine. Can you send full or subset of some1.fastq file for testing? Also, can you try with one other file? May be use the test file available here. Let me know how this goes.

Thank you,
Arun

shashankpritam commented

Hi! Yes, The test file works as explained in the quick start docs.

Here's what I have been running the miRge over -

@SRR21820251.1 1 length=99
NGAGATTTTATTTCGGCAGATCATGACGATATCGCAAACTCGTGAACCTTACCACGCGTCGTGTGTGTACGTGTGTGTATATATAGCATACGTAAACTA
+SRR21820251.1 1 length=99
#FFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFFFFFFFFFF,FFFFF
@SRR21820251.2 2 length=100
NTTTCTTTGGACCTGTCTTACCTTGAGGATCATATGGAAGCATAATCTTCACTTTGATCCCAAGAACACCTTGTCGAAGAAGCACATGACGGGTTGCAGT
+SRR21820251.2 2 length=100
#FFFFFFFFFFFFFFFFFFFFFFFFFFFF:FFFF,FFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFFFFFFFFFFFFF
@SRR21820251.3 3 length=100
NGCATGTATTTCATTATACTTTTCACTTGTATCAAAGGTAGCTCTTCGTTCAGCGACGAACATTTTCCTGCTTCCATTAACATCTTCATTTGTAGCCTGT
umassemb$ head SRR21820251_2.fastq
@SRR21820251.1 1 length=100
ATGTACTGACGTCTGGCTAAGGAGAGAAAGGGTGAGAGCGTATACAATCAATTAGTTTTTTAGTTTACGTATGCTATATATACACACACGTACACACACG
+SRR21820251.1 1 length=100
FFFFFFFFFFFFFFFFFFFFFFF:FFFFFFF,FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF:FFFFF:FF:FFFF:FFFFFFFFFFFFF
@SRR21820251.2 2 length=100
TGTTGTTAGCGGTAAATTGCGTGGACAAAGAGCAAAATCAATGAAATTCGTTGATGGATTGATGATTCACTCTGGTGAACCCACCAATGAATATGTTAAT
+SRR21820251.2 2 length=100
FFF:FFFFFFFF,:FFFFFF:FFFFFFFFFFFFFFFFFFFFF,FFFFFFF,:FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF,FFFFFFFFFFFFF
@SRR21820251.3 3 length=100
CAACAGTTGTCGGCAGGTTCTTACGGATGTTCTACACATAGACTTATACCAAGGAAACAGGCTACAAATGAAGATGTTAATGGAAGCAGGAAAATGTTCG
umassemb$ head out_SRR21820251_1.fastq
@SRR21820251.1 1 length=99
NGAGATTTTATTTCGGCAGATCATGACGATATCGCAAACTCGTGAACCTTACCACGCGTCGTGTGTGTACGTGTGTGTATATATAGCATACGTAAACTA
+SRR21820251.1 1 length=99
#FFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFFFFFFFFFF,FFFFF
@SRR21820251.2 2 length=100
NTTTCTTTGGACCTGTCTTACCTTGAGGATCATATGGAAGCATAATCTTCACTTTGATCCCAAGAACACCTTGTCGAAGAAGCACATGACGGGTTGCAGT
+SRR21820251.2 2 length=100
#FFFFFFFFFFFFFFFFFFFFFFFFFFFF:FFFF,FFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFF:FFFFFFFFFFFFFFFFFFFFFFFFFFFFFF
@SRR21820251.3 3 length=100
NGCATGTATTTCATTATACTTTTCACTTGTATCAAAGGTAGCTCTTCGTTCAGCGACGAACATTTTCCTGCTTCCATTAACATCTTCATTTGTAGCCTGT
umassemb$ head out_SRR21820251_2.fastq
@SRR21820251.1 1 length=100
ATGTACTGACGTCTGGCTAAGGAGAGAAAGGGTGAGAGCGTATACAATCAATTAGTTTTTTAGTTTACGTATGCTATATATACACACACGTACACACACG
+SRR21820251.1 1 length=100
FFFFFFFFFFFFFFFFFFFFFFF:FFFFFFF,FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF:FFFFF:FF:FFFF:FFFFFFFFFFFFF
@SRR21820251.2 2 length=100
TGTTGTTAGCGGTAAATTGCGTGGACAAAGAGCAAAATCAATGAAATTCGTTGATGGATTGATGATTCACTCTGGTGAACCCACCAATGAATATGTTAAT
+SRR21820251.2 2 length=100
FFF:FFFFFFFF,:FFFFFF:FFFFFFFFFFFFFFFFFFFFF,FFFFFFF,:FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF,FFFFFFFFFFFFF
@SRR21820251.3 3 length=100
CAACAGTTGTCGGCAGGTTCTTACGGATGTTCTACACATAGACTTATACCAAGGAAACAGGCTACAAATGAAGATGTTAATGGAAGCAGGAAAATGTTCG

Data is from NCBI - RNA-seq of megachile rotundata

This could be an issue related with the data.

Info - the out_files are QC fastq files after runnning FASTP over the original.

Thank you,
Shashank

arunhpatil commented

Hi @shashankpritam,

Yes. This accession SRR21820251 is paired-end derived RNA-seq data, I downloaded 'SRR21820251' it to see if there are any miRNAs in them and my grep search of most abundant miRNA let-7a is missing, which means there are no miRNAs in the data. miRge3.0 does not support paired-end reads, they need to be processed as single end (analyzing each end seperately). The reason is that, miRNAs are short sequences of ~25 nucleotides and paired-end experiments are not ideal.

Can you run the following accession with your pipeline, this is from the same source but specific to miRNA-sequencing:
SRR21851723

Having no miRNA reads has caused the issue. I hope this helps.

Thank you,
Arun.

shashankpritam commented

Thank you very much @arunhpatil! Yes, it works. I got the output files with SRR21851723 -

annotation.report.csv  annotation.report.html  index_data.js  mapped.csv  miR.Counts.csv  miRge3_visualization.html  miR.RPM.csv  run.log  unmapped.csv

Thanks again for the help and guidance,
Shashank

  • 👍1