a process with conditional channel

[Damian Loska]

Hey, is it possible to code a single process with different channels?

nLines_per_WES_bed = ((file(params.bed_flank_100_file).readLines().size()+50)/30).toInteger()

Channel.fromPath( params.bed_flank_100_file ).splitText( by:nLines_per_WES_bed, file:params.yy_splits_dir)
  .into { WESbedSplitsA; WESbedSplitsB; WESbedSplitsC }


Channel.fromPath( params.bed_wgs_file ).splitText( by:1, file:params.yy_splits_dir)
  .into { WGSbedSplitsA; WGSbedSplitsB; WGSbedSplitsC }

wgs=true
process test {
    echo true

    input:
        file(x) from ThisChannel

    script:
        if wgs:
            ThisChannel = WGSbedSplitsA
        else:
            ThisChannel = WESbedSplitsA


"""
echo $x

"""
}

I want a HaplotypeCaller with a different .bed content, depends whether it is WGS or WES.

Thanks

Damian

[drhp...@gmail.com]

Hi Damian! It might be easier to move that sort of logic outside of the process to make it more generically usable. Something like below:

wgs = true

if (wgs) {

    nLines_per_WES_bed = ((file(params.bed_flank_100_file).readLines().size()+50)/30).toInteger()

    Channel

      .fromPath( params.bed_flank_100_file ).splitText( by:nLines_per_WES_bed, file:params.yy_splits_dir)

      .set { bedSplits }

} else {

    Channel.fromPath( params.bed_wgs_file ).splitText( by:1, file:params.yy_splits_dir)

      .set { bedSplits }

}

And then give the process a single channel called bedSplits.

[Damian Loska]

Yeah, that's what I finally did, but such an approach sets the .bed channel at the very beginning of the script, which means that I can run a WES pipeline, or a WGS pipeline. I wanted to make a "mixed" WES/WGS pipeline, so it would be processed at once, switching WES.bed or WGS.bed depending on the _name_ coded in sample name :)

(xxx_WGS_yyy_zzz_bbb.1.fastq.gz) 

--
You received this message because you are subscribed to a topic in the Google Groups "Nextflow" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/nextflow/Yk73IGgPdJI/unsubscribe.
To unsubscribe from this group and all its topics, send an email to nextflow+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/nextflow/e10849f2-32bf-4b30-9a81-6e24e11e8b9fn%40googlegroups.com.

[drhp...@gmail.com]

Ah, so in that case you could try to always stage the intervals files for both types of prep. Might also be nicer to encode this sort of sample metadata in a Groovy map. Some pseudo-code in DSL2 below:

WESbedSplits = []

if (params.bed_flank_100_file) {

    nLines_per_WES_bed = ((file(params.bed_flank_100_file).readLines().size()+50)/30).toInteger()

    Channel

      .fromPath( params.bed_flank_100_file ).splitText( by:nLines_per_WES_bed, file:params.yy_splits_dir)

      .set { WESbedSplits }

}

WGSbedSplits = []

if (params.bed_wgs_file) {

    Channel.fromPath( params.bed_wgs_file ).splitText( by:1, file:params.yy_splits_dir)

      .set { WGSbedSplits }

}

process test {

    input:

    tuple val(meta), path(fastq)

    path wes_intervals

    path wgs_intervals

    script:

    if (meta.prep == 'wgs' && wgs_intervals) {

        wxs_opts = "--intervals $wgs_intervals"

    } else if (meta.prep != 'wgs' && wes_intervals) {

        wxs_opts = "--intervals $wes_intervals"

    }

    """

    echo $meta.prep

    echo $wxs_opts

    """

}