ERROR ~ Error executing process > 'NFCORE_VIRALRECON:NANOPORE:ARTIC_MINION'

[alethomas]

Description of the bug

When running nf-core/viralrecon, the medaka minion throws an error. Seems the model cannot be downloaded via the medaka minion script. scm file with github credentials is present, internet connection is available.

Command used and terminal output

nextflow run nf-core/viralrecon -qs 4 -revision 2.6.0 -profile docker 
--input results/benchmarking/nf-core-viralrecon/nanopore/sample-sheets/2_ont/sample_sheet.csv --sequencing_summary data/RAW/ONT_batch2/sequencing_summary.txt 
--fastq_dir resources/benchmarking/data/nf-core-viralrecon/2_ont/fastq_pass 
--fast5_dir resources/benchmarking/data/nf-core-viralrecon/2_ont/fast5_pass 
--platform nanopore 
--genome 'MN908947.3' 
--primer_set_version 3 
--artic_minion_caller medaka 
--artic_minion_medaka_model r941_min_fast_g303 
--outdir results/benchmarking/nf-core-viralrecon/nanopore/medaka/2_ont

Command error:
  [M::mm_idx_stat::0.007*3.40] distinct minimizers: 5587 (99.93% are singletons); average occurrences: 1.004; average spacing: 5.332; total length: 29903
  [M::worker_pipeline::17.215*4.14] mapped 365148 sequences
  [M::main] Version: 2.24-r1122
  [M::main] CMD: minimap2 -a -x map-ont -t 12 ./primer-schemes/nCoV-2019/V3/nCoV-2019.reference.fasta 2_ont.fastq.gz
  [M::main] Real time: 17.218 sec; CPU: 71.326 sec; Peak RSS: 0.489 GB
  Traceback (most recent call last):
    File "/usr/local/lib/python3.8/site-packages/medaka/medaka.py", line 35, in __call__
      model_fp = medaka.models.resolve_model(val)
    File "/usr/local/lib/python3.8/site-packages/medaka/models.py", line 66, in resolve_model
      raise DownloadError(
  medaka.models.DownloadError: The model file for r941_min_fast_g303 is not already installed and could not be downloaded. Check you are connected to the internet and try again.

Relevant files

nextflow.log

System information

nextflow version 23.04.1.5866
Linux server
locally executed in docker container
nf-core/viralrecon 2.6.0

[svarona]

Sorry for the late reply @alethomas ! We will have a closer look to this to see what happens, but a workaround we do for the CI tests is downloading the model before running the pipeline trough: wget https://github.com/nanoporetech/medaka/raw/master/medaka/data/r941_min_high_g360_model.hdf5

This might temporarily solve the problem.