main.nf

/*
 * Copyright (c) 2023. The University of Texas Southwestern Medical Center
 *
 *   This file is part of the BioHPC Workflow Platform
 *
 * This is a workflow package to run cellranger count on fastq files from single cell RNA data.
 *
 * @authors
 * John Lafin
 *
 */


// Parameters for input values
params.sample = "Brain_Tumor_3p_LT"
params.fastq = "${projectDir}/../test_data/Brain_Tumor_3p_LT_fastqs"
params.reference = "hg38"
params.expectCells = 0
params.chemistry = "auto"
params.introns = true
params.noBam = false
params.outDir = "${projectDir}/output"

// Run cellranger count
process cr_count {
    publishDir "${outDir}", mode: 'copy'
    queue "256GB,256GBv1"
    module 'cellranger/7.1.0'
    module 'singularity/3.9.9'

    input: 
        val sample
        path ref
        path fastq
        val expectCells
        val chemistry
        val introns
        val noBam
        path outDir

    output:
        tuple val(sample), path("**/outs/**")

    script:
        if( expectCells == 0 )
            if( noBam )
            """
            cellranger count --id=$sample \
                            --transcriptome=$ref \
                            --fastqs=. \
                            --sample=$sample \
                            --chemistry=$chemistry \
                            --include-introns=$introns \
                            --no-bam 
            """
            else
            """
            cellranger count --id=$sample \
                            --transcriptome=$ref \
                            --fastqs=. \
                            --sample=$sample \
                            --chemistry=$chemistry \
                            --include-introns=$introns
            """

        else if( expectCells > 0)
            if( noBam )
            """
            cellranger count --id=$sample \
                            --transcriptome=$ref \
                            --fastqs=. \
                            --sample=$sample \
                            --expect-cells=$expectCells \
                            --chemistry=$chemistry \
                            --include-introns=$introns \
                            --no-bam 
            """
            else
            """
            cellranger count --id=$sample \
                            --transcriptome=$ref \
                            --fastqs=. \
                            --sample=$sample \
                            --expect-cells=$expectCells \
                            --chemistry=$chemistry \
                            --include-introns=$introns
            """
}

// Download reference genome if missing
process get_reference {
    queue "super"
    module 'singularity/3.9.9'

    input:
        val ref_name

    output:
        path ("refdata*", type: "dir")

    script:
        if( ref_name=="hg38" ) {
            """
            wget -q https://cf.10xgenomics.com/supp/cell-exp/refdata-gex-GRCh38-2020-A.tar.gz
            tar -zxf refdata-gex-GRCh38-2020-A.tar.gz
            rm refdata-gex-GRCh38-2020-A.tar.gz
            """
        }
        else if( ref_name=="mm10" ) {
            """
            wget -q https://cf.10xgenomics.com/supp/cell-exp/refdata-gex-mm10-2020-A.tar.gz
            tar -zxf refdata-gex-mm10-2020-A.tar.gz
            rm refdata-gex-mm10-2020-A.tar.gz
            """
        }
}

workflow {
    // Select reference genome
    if (file(params.reference).exists()) {
        ref = file(params.reference)
    }
    else if (params.reference == "hg38") {
        ref = file("/project/apps_database/cellranger/refdata-gex-GRCh38-2020-A")
    }
    else if (params.reference == "mm10") {
        ref = file("/project/apps_database/cellranger/refdata-gex-mm10-2020-A")
    }
    else {
        ref = file("missing_reference")
    }

    // Download reference if missing
    if( ref.isEmpty() ) {
        ref = get_reference(params.reference)
    }

    // Define channels from variables
    sample = Channel.value(params.sample)
    fastq = Channel.fromPath(params.fastq).collect()
    expectCells = Channel.value(params.expectCells)
    chemistry = Channel.value(params.chemistry)
    introns = Channel.value(params.introns)
    noBam = Channel.value(params.noBam)
    outDir = Channel.value(params.outDir)

    // Run cellranger count
    cr_count(sample, ref, fastq, expectCells, chemistry, introns, noBam, outDir)
}