Develop

[thorellk]

I guess we should then merge this to master so that we get all those changes included?

[boulund]

I think we're very close to completing this now! Has anyone encountered any odd error messages that were not handled gracefully?

I'll make one last check to see if version numbers of correct everywhere.

[boulund]

Sorry for putting all these commit directly on the develop branch, I started with a very small change, but found lots of minor things I wanted to improve across the documentation.

[emilio-r]

odd error messages that were not handled gracefully

As comes to mind, some errors that affect the end user may be:

• Not having a profile assigned. Most users would either want to have local or Docker so maybe we should then say that if none is specifically assigned, local will be run as default?
• Not assigning a Kraken2 database and that this will skip that module.
• Running an outdated Nextflow version and getting the stub-error, or some other related error.
• Not having SignalP installed (but this will be applied in a later version)
• That the BACTpipe arguments are incorrectly entered (paths not in quotes, empty, etc.)

[boulund]

• Not having a profile assigned. Most users would either want to have local or Docker so maybe we should then say that if none is specifically assigned, local will be run as default?

This is mentioned in the docs, in this section: https://bactpipe.readthedocs.io/en/develop/running.html#running-bactpipe

Additionally, if running a cluster profile that requires a project name, BACTpipe throws an error message. E.g. if the Rackham profile is selected but no --project is given. This is also mentioned here: https://bactpipe.readthedocs.io/en/develop/running.html#profiles.

Example:

$ nextflow run ~/bactpipe/bactpipe.nf --kraken2_db /db/kraken2/minikraken2_v2_8GB_201904_UPDATE/ --reads 'raw_data/V300034817/subset/*_{1,2}.fq.gz' -profile rackham
N E X T F L O W  ~  version 21.04.0
Launching `/ceph/home/boulund/bactpipe/bactpipe.nf` [shrivelled_knuth] - revision: f2910af028
============================================================
                          BACTpipe
                       Version 3.1.0
          Bacterial whole genome analysis pipeline
              https://bactpipe.readthedocs.io
============================================================
BACTpipe requires that you set the 'project' parameter when running the rackham profile.
Specify --project <project_name> on the command line, or add it to a custom configuration file.Refer to the official docs for more information.

• Not assigning a Kraken2 database and that this will skip that module.

BACTpipe prints a warning about it, and it is mentioned in the docs and implied in the command line help.

Example:

$ nextflow run ~/bactpipe/bactpipe.nf --reads 'raw_data/V300034817/subset/*_{1,2}.fq.gz'
N E X T F L O W  ~  version 21.04.0
Launching `/ceph/home/boulund/bactpipe/bactpipe.nf` [jolly_volhard] - revision: f2910af028
============================================================
                          BACTpipe
                       Version 3.1.0
          Bacterial whole genome analysis pipeline
              https://bactpipe.readthedocs.io
============================================================
            Running with the following settings:
                       reads: raw_data/V300034817/subset/*_{1,2}.fq.gz
                     project:
              clusterOptions: false
                  output_dir: BACTpipe_results
                  kraken2_db:
          kraken2_confidence: 0.5
      kraken2_min_proportion: 1.00
               shovill_depth: 100
               shovill_kmers: 31,33,55,77,99,127
              shovill_minlen: 500
               prokka_evalue: 1e-09
              prokka_kingdom: Bacteria
            prokka_reference:
      prokka_signal_peptides: true
          keep_trimmed_fastq: false
         keep_shovill_output: false
============================================================
WARN: No Kraken2 database specified. Use --kraken2_db /path/to/db to use Kraken2 to classify samples and determine gram stain.
[-        ] process > FASTP             -
[-        ] process > CLASSIFY_TAXONOMY -
[-        ] process > SHOVILL           -
... etc

• Running an outdated Nextflow version and getting the stub-error, or some other related error.

Not sure about this one. We do have this line here, but not sure if that is actually enforced or what kind of error message is printed if an old version is used: https://github.com/ctmrbio/BACTpipe/blob/develop/nextflow.config#L8

• Not having SignalP installed (but this will be applied in a later version)

This issue is still open I think and has been pushed to our next release: #148

• That the BACTpipe arguments are incorrectly entered (paths not in quotes, empty, etc.)

This is kind of fixed, but still has a minor bug (the workflow completed "without errors" even if it failed 😄 ...

$ nextflow run ~/bactpipe/bactpipe.nf
N E X T F L O W  ~  version 21.04.0
Launching `/ceph/home/boulund/bactpipe/bactpipe.nf` [maniac_mclean] - revision: f2910af028
============================================================
                          BACTpipe
                       Version 3.1.0
          Bacterial whole genome analysis pipeline
              https://bactpipe.readthedocs.io
============================================================
            Running with the following settings:
                     project:
              clusterOptions: false
                  output_dir: BACTpipe_results
                       reads:
                  kraken2_db:
          kraken2_confidence: 0.5
      kraken2_min_proportion: 1.00
               shovill_depth: 100
               shovill_kmers: 31,33,55,77,99,127
              shovill_minlen: 500
               prokka_evalue: 1e-09
              prokka_kingdom: Bacteria
            prokka_reference:
      prokka_signal_peptides: true
          keep_trimmed_fastq: false
         keep_shovill_output: false
============================================================
WARN: No Kraken2 database specified. Use --kraken2_db /path/to/db to use Kraken2 to classify samples and determine gra
m stain.
Path string cannot be empty


Cannot find any reads matching: ''
Did you specify --reads 'path/to/*_{1,2}.fastq.gz'? (note the single quotes)
Specify --help for a summary of available commands. Refer to the official docs for more information.



  Example usage:
    nextflow run ctmrbio/BACTpipe --reads '*_R{1,2}.fastq.gz'

  Mandatory arguments:
    --reads                 Path to input data (must be surrounded with single quotes).

  Optional arguments:
    --kraken2_db            Path to Kraken2 database for taxonomic classification and gram
                            stain determination (recommended!).
    --keep_trimmed_fastq    Save trimmed fastq files in output directory (default: false).
    --keep_shovill_output   Save all shovill output in output directory (default: false).

  Output options:
    --output_dir            Output directory, where results will be saved
                            (default: BACTpipe_results).

  Refer to the online manual for more information and all available options:
             https://bactpipe.readthedocs.io

============================================================
         BACTpipe workflow completed without errors
Oops... Pipeline execution stopped with the following message: Path string cannot be empty
Path string cannot be empty

[boulund]

I'm happy with how this works now. Let me know if you think otherwise. We can merge later today or tomorrow unless there is anything else.

[abhi18av]

I'm happy with the changes for this release as well 👍

[emilio-r]

I am also happy with the way this looks and behaves, so I reaffirm my previous approval of this merge.