edgeR_mirna error

[mshulewitz]

Running smrnaseq on an AWS EC2 (Ubuntu 18.04, 64 GB memory, 16 cpu) with the following command line:

nextflow run nf-core/smrnaseq -r 1.0.0 -profile docker \
--reads 'sample1.*' \
--genome 'GRCh37' \
--max_memory '60.GB' \
 --outdir smrnaseq_output \
--protocol cats

gives the following error:

Error executing process > 'edgeR_mirna'

Caused by:
  Process `edgeR_mirna` terminated with an error exit status (1)

Command executed:

  edgeR_miRBase.r Kit4RNA3.mature.stats Kit4RNA3.hairpin.stats

Command exit status:
  1

Command output:
  $mature
  [1] "sample1.mature.stats"

  $hairpin
  [1] "sample1.hairpin.stats"

Command error:
  WARNING: Your kernel does not support swap limit capabilities or the cgroup is not mounted. Memory limited without swap.
  Loading required package: limma
  Loading required package: edgeR
  Loading required package: statmod
  Loading required package: data.table
  Loading required package: gplots

  Attaching package: ‘gplots’

  The following object is masked from ‘package:stats’:

      lowess

  Error in apply(data, 1, function(row) all(row == 0)) :
    dim(X) must have a positive length
  Execution halted

Running edgeR_miRBase.r locally also gives same error.

[lpantano]

Hi,

I think something is wrong with the input. Can you go into the working folder and check the input files, if they are not empty?

maybe try to run the R script in an interactive version helps more as well.

[mshulewitz]

I pulled the test fastq and reference files using the paths in test.config and rerun with the following command line and was successful:

nextflow run nf-core/smrnaseq -r 1.0.0 -profile docker \
--reads 'fastq_repo/sample_*' \
--max_memory '60.GB' \
--mature 'smrnaseq_ref/mature.fa' \
--hairpin 'smrnaseq_ref/hairpin.fa' \
--bit_index 'reference/genome' \
--fasta 'smrnaseq_ref/genes.gtf' \
--gtf 'smrnaseq_ref/hsa.gff3' \
--mirtrace_species 'hsa' \
--outdir smrnaseq_output

However, if I only process a single fastq (eg: --reads 'fastq_repo/sample_1.fastq.gz') I receive the error message. In this case, the input files sample_1.hairpin.stats and sample_1.mature.stats are in the working directory for edgeR_mirna process.

[lpantano]

Ah, ok, I see the error, it is only one sample the problem… I would need to take a look into the R code exactly to know why. (Sorry for the delay I have been away until now.)

[rakeshponnala]

I'm getting the same error.. was this issue fixed... I'm trying to run 98 samples.. the pipeline runs great.. but at the end throws similar error message:

Error executing process > 'edgeR_mirna'

Caused by:
Process edgeR_mirna terminated with an error exit status (1)

Command executed:

Command error:
Loading required package: limma
Loading required package: edgeR
Loading required package: statmod
Loading required package: data.table
Loading required package: gplots

Attaching package: ‘gplots’

The following object is masked from ‘package:stats’:

  lowess

Error in if (nr_keep > 0) { : argument is of length zero
Execution halted

[lpantano]

Hi, what version of the pipeline are you using?

[rakeshponnala]

r 1.1.0 is the version I am using...

[lpantano]

could you try running the dev branch instead?

[apeltzer]

Goign to close this as the issue is very likely not there anymore in the current dev and soon release - please retest again 👍🏻