nextflow.config

/*
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    nf-core/dragenflow Nextflow config file
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    Default config options for all compute environments
----------------------------------------------------------------------------------------
*/

// Global default params, used in configs
params {

    // inputs THESE NEED TO BE SET
    input                       = null
    outdir                      = null
    demux_outdir                = null
    target_bedfile              = null
    tracedir                    = "${params.outdir}/pipeline_info"

    // compute1 cluster options
    user_group                   = "compute-dspencer"
    queue                        = "general"
    job_group_name               = "/dspencer/nextflow"

    // compute2 cluster options
    slurm_partition              = "condo-dspencer"
    slurm_account                = "compute2-dspencer"
    apptainer_cache            = '/scratch2/fs1/dspencer/singularity_cache'

    // workflow parameter. This sets the analysis mode.
    // It is set with the analysis profiles (see below).
    workflow                    = null

    // Can add aribitrary arguments, if needed.
    extra_dragen_args           = null

    // Extra Dragen options
    alignment_file_format      = 'CRAM'
    mark_duplicates            = true
    pangenome_reference        = false
    variant_caller             = false
    sv_caller                  = false
    cnv_caller                 = false
    umi                        = false
    illumina_conversion        = false
    readfamilysize             = 3
    solid_tumor                = false
    liquid_tumor               = false
    downsample_rna             = null
    dux4caller                 = false
    dragen_cnv_filter_length   = null //500000
    dragen_cnv_merge_distance  = null //2000000
    sv_annotation_distance     = null //20000
    max_filter_sv_length       = null //150000000

    // DRAGEN alignment options
    intermediate_dir                  = "/staging/intermediate-results-dir"
    refdir                            = "/storage2/fs1/dspencer/Active/shared/refdata/hg38/dragen_hg38v4.3.6"
    dbsnp                             = "/storage2/fs1/dspencer/Active/shared/refdata/hg38/dragenfiles/dbsnp.vcf.gz"
    methylation_refdir                = null
    adapter1                          = "${projectDir}/assets/data/dragen_adapter1.fa"
    adapter2                          = "${projectDir}/assets/data/dragen_adapter2.fa"
    annotation_gtf                    = null
    transcript_table                  = null
    cnv_population_vcf                = null
    snv_noisefile                     = null
    sv_noisefile                      = null
    hotspot_vcf                       = null
    hotspot_bed                       = null
    dragen_tandem_dup_hotspots        = null
    
    // References
    fasta                             = "/storage2/fs1/dspencer/Active/shared/refdata/hg38/sequence/hg38_mgi_patch.fa"
    cram_reference                    = "/storage2/fs1/dspencer/Active/shared/refdata/hg38/sequence/hg38_mgi_patch.fa"
    
    // VEP and Nirvana annotation dirs
    vepcache                          = "/storage2/fs1/dspencer/Active/shared/refdata/hg38/VEP113_cache"
    nirvana_path                      = "/storage2/fs1/dspencer/Active/shared/refdata/hg38/dragenfiles/nirvana_annotation_data_454"
    nirvana_assembly                  = "GRCh38"
    use_nirvana                       = true

    // cytobanbs
    cytobands                         = "${projectDir}/assets/data/hg38.cytoBandIdeo.bed.gz"

    local_dragen_container            = "apldx/oracle8-dragen-4.3.6"

    // for testing
    run_dragen = true

    // For building dragen hash
    build_rna_hash_table            = true
    build_hla_hash_table            = true
    build_cnv_hash_table            = false
    build_legacy_cnv_hash_table     = true
    build_methylation_hash_table    = false
    
    // TODO nf-core: Specify your pipeline's command line flags
    // Input options
    // References
    genome                     = null
    igenomes_base              = 's3://ngi-igenomes/igenomes'
    igenomes_ignore            = true
    
    // MultiQC options
    multiqc_config             = null
    multiqc_title              = null
    multiqc_logo               = null
    max_multiqc_email_size     = '25.MB'
    multiqc_methods_description = null

    // Boilerplate options
    outdir                     = null
    publish_dir_mode           = 'copy'
    email                      = null
    email_on_fail              = null
    plaintext_email            = false
    monochrome_logs            = false
    hook_url                   = null
    help                       = false
    version                    = false

    // Config options
    config_profile_name        = null
    config_profile_description = null
    custom_config_version      = 'master'
    custom_config_base         = "https://raw.githubusercontent.com/nf-core/configs/${params.custom_config_version}"
    config_profile_contact     = null
    config_profile_url         = null
    

    // Max resource options
    // Defaults only, expecting to be overwritten
    max_memory                 = '128.GB'
    max_cpus                   = 16
    max_time                   = '240.h'

    // Schema validation default options
    validationFailUnrecognisedParams = false
    validationLenientMode            = false
    validationSchemaIgnoreParams     = 'genomes,igenomes_base'
    validationShowHiddenParams       = false
    validate_params                  = true

}

// Load igenomes.config if required
if (!params.igenomes_ignore) {
    includeConfig 'conf/igenomes.config'
} else {
    params.genomes = [:]
}

// Nextflow plugins
plugins {
    id 'nf-schema@2.7.3'
    id 'nf-amazon@3.4.1'
}

// Load base.config by default for all pipelines
includeConfig 'conf/base.config'

profiles {
    
    // Dragen analysis profiles
    alignonly      { includeConfig 'conf/alignonly.config' }
    germline       { includeConfig 'conf/germline.config' }
    somaticheme    { includeConfig 'conf/somaticheme.config' }
    somaticsolid   { includeConfig 'conf/somaticsolid.config' }
    tumoronlyheme  { includeConfig 'conf/tumoronlyheme.config' }
    tumoronlysolid { includeConfig 'conf/tumoronlysolid.config' }
    idtumi         { includeConfig 'conf/idtumi.config' }
    rnaseq         { includeConfig 'conf/rnaseq.config' }
    fivebase       { includeConfig 'conf/5base.config' }

    // For latest DRAGEN version
    dragenaws   { includeConfig 'conf/dragenAWS454.config' }

    // Specific Dragen AWS versions
    dragenAWS454   { includeConfig 'conf/dragenAWS454.config' }
    dragenAWS446   { includeConfig 'conf/dragenAWS446.config' }

    // Full AWS mode--still need to load a dragen config
    dhslabaws      { includeConfig 'conf/dhslabaws.config' }

    test { includeConfig 'conf/test.config' } // this run demux on first tile only
    stub { includeConfig 'conf/stub.config' } // local executor for testing
    
    ris {
        executor.queueSize        = 20
        executor.submitRateLimit  = '1/1sec'
        process {
            executor                  = "lsf"
            clusterOptions            =   { "-a 'docker(${task.container})' -q ${params.queue} -G ${params.user_group} -g ${params.job_group_name}" }
        }
    }

    ris2 {
        apptainer.enabled       = true
        apptainer.autoMounts    = true
        apptainer.cacheDir      = "${params.apptainer_cache}"
        executor.queueSize       = 100
        executor.submitRateLimit = '1/1sec'
        process {
            executor             = "slurm"
            queue                = "${params.slurm_partition}"
            clusterOptions       = "-A ${params.slurm_account}"
        }
    }
    
    dragen4 {
        process {
            withLabel: 'dragen' {
                ext.dragen_container = "${params.local_dragen_container}"            
                ext.dragen_path = "/opt/dragen/4.3.6"
                queue = { "dragen-4" }
                memory  = 300.GB
                cpus = 30
                time = 16.h                
                clusterOptions = { "-a 'gtac-mgi-dragen(${task.container})' -m compute1-dragen-4 -G ${params.user_group} -g ${params.job_group_name} -env 'all, LSF_DOCKER_DRAGEN=y'" }
            }
        }
    }
    dragen5 {
        process {
            withLabel: 'dragen' {
                ext.dragen_container = "${params.local_dragen_container}"
                ext.dragen_path = "/opt/dragen/4.3.6"
                queue = { "dragen-5" }
                memory  = 300.GB
                cpus = 30
                time = 16.h                
                clusterOptions = { "-a 'gtac-mgi-dragen(${task.container})' -m compute1-dragen-5 -G ${params.user_group} -g ${params.job_group_name} -env 'all, LSF_DOCKER_DRAGEN=y'" }
            }
        }
    }
}

// Set default registry for Apptainer, Docker, Podman and Singularity independent of -profile
// Will not be used unless Apptainer / Docker / Podman / Singularity are enabled
// Set to your registry if you have a mirror of containers
apptainer.registry   = 'quay.io'
docker.registry      = 'quay.io'
podman.registry      = 'quay.io'
singularity.registry = 'quay.io'

// Export these variables to prevent local Python/R libraries from conflicting with those in the container
// The JULIA depot path has been adjusted to a fixed path `/usr/local/share/julia` that needs to be used for packages in the container.
// See https://apeltzer.github.io/post/03-julia-lang-nextflow/ for details on that. Once we have a common agreement on where to keep Julia packages, this is adjustable.

env {
    PYTHONNOUSERSITE = 1
    R_PROFILE_USER   = "/.Rprofile"
    R_ENVIRON_USER   = "/.Renviron"
    JULIA_DEPOT_PATH = "/usr/local/share/julia"
}

// Capture exit codes from upstream processes when piping
process.shell = ['/bin/bash', '-euo', 'pipefail']

def trace_timestamp = new java.util.Date().format( 'yyyy-MM-dd_HH-mm-ss')
timeline {
    enabled = true
    file    = "${params.outdir}/pipeline_info/execution_timeline_${trace_timestamp}.html"
}
report {
    enabled = true
    file    = "${params.outdir}/pipeline_info/execution_report_${trace_timestamp}.html"
}
trace {
    enabled = true
    file    = "${params.outdir}/pipeline_info/execution_trace_${trace_timestamp}.txt"
}
dag {
    enabled = true
    file    = "${params.outdir}/pipeline_info/pipeline_dag_${trace_timestamp}.html"
}

manifest {
    name            = 'dhslab/dragenflow'
    author          = """David Spencer"""
    homePage        = 'https://github.com/dhslab/dragenflow'
    description     = """Run multple dragen workflows (5mc,rna,tumor_normal)"""
    mainScript      = 'main.nf'
    nextflowVersion = '!>=25.10.0'
    version         = '1.0dev'
    doi             = ''
}

// Load modules.config for DSL2 module specific options
includeConfig 'conf/modules.config'