nextflow.config

/*
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    genomic-medicine-sweden/nallo Nextflow config file
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    Default config options for all compute environments
----------------------------------------------------------------------------------------
*/

// Global default params, used in configs
params {

    // Preset
    preset                         = 'revio'

    // Input options
    input                          = null
    cadd_resources                 = null
    cadd_prescored_indels          = null
    fasta                          = null
    filter_variants_hgnc_ids       = null
    methbat_regions                = null
    par_regions                    = null
    paraphrase_rules               = null
    tandem_repeats                 = null
    str_bed                        = null
    stranger_repeat_catalog        = null
    genmod_reduced_penetrance      = null
    genmod_score_config_snvs       = null
    genmod_score_config_svs        = null
    echtvar_snv_databases          = null
    svdb_sv_databases              = null
    variant_consequences_snvs      = null
    variant_consequences_svs       = null
    vep_cache                      = null
    vep_plugin_files               = null
    cnv_expected_xy_cn             = null
    cnv_expected_xx_cn             = null
    cnv_excluded_regions           = null
    peddy_sites                    = null
    somalier_sites                 = null
    strdrop_training_set_json      = null
    gens_baf_positions             = null
    gens_panel_of_normals_female   = null
    gens_panel_of_normals_male     = null
    gens_coverage_bins             = null

    // Regions options
    target_regions                 = null
    modkit_call_regions            = params.target_regions
    snv_call_regions               = params.target_regions
    sv_call_regions                = params.target_regions

    // QC regions options
    qc_regions                     = params.target_regions
    mosdepth_regions               = params.qc_regions
    sambamba_regions               = params.qc_regions

    // Output options
    publish_unannotated_family_svs = false
    alignment_output_format        = 'bam'
    bigwig_modcodes                = params.preset == 'ONT_R10' ? 'h,m' : 'm'
    create_maf_track               = true
    create_hificnv_maf_track       = params.create_maf_track
    create_sawfish_maf_track       = params.create_maf_track

    // Preset
    preset                         = 'revio'

    // Chromograph plotting options
    chromograph_af_tag             = null
    bcftools_roh_af_tag            = params.chromograph_af_tag
    rhocallviz_af_tag              = params.chromograph_af_tag
    rhocallviz_min_af              = 0.0001
    rhocallviz_min_qual            = 10
    plot_chromograph_coverage      = true
    plot_chromograph_autozygosity  = (params.bcftools_roh_af_tag && params.rhocallviz_af_tag && !params.skip_snv_annotation) // assume if these are set, the user has set correct AF-tags

    // Skip parts of the workflow
    skip_alignment                 = false
    skip_annotate_paralogs         = params.preset == 'ONT_R10' ? true : false
    skip_call_paralogs             = params.preset == 'ONT_R10' ? true : false
    skip_chromograph               = !(params.plot_chromograph_autozygosity && params.plot_chromograph_coverage)
    skip_genome_assembly           = false
    skip_methylation_calling       = params.preset == 'pacbio' ? true : false
    skip_peddy                     = false
    skip_phasing                   = false
    skip_qc                        = false
    skip_rank_variants             = false
    skip_repeat_calling            = false
    skip_repeat_annotation         = params.preset == 'ONT_R10' ? true : false
    skip_sambamba_depth            = !params.sambamba_regions
    skip_snv_calling               = false
    skip_snv_annotation            = false
    skip_sv_annotation             = false
    skip_sv_calling                = false
    skip_prepare_gens_input        = false

    deepvariant_model_type         = params.preset == 'ONT_R10' ? 'ONT_R104' : 'PACBIO'
    hifiasm_preset                 = params.preset == 'ONT_R10' ? '--ont' : ''
    methbat_male_label             = 'MALE'
    methbat_female_label           = 'FEMALE'
    minimap2_read_mapping_preset   = params.preset == 'ONT_R10' ? 'lr:hq' : 'map-hifi'
    modkit_modified_bases          = params.preset == 'ONT_R10' ? '5mC 5hmC' : '5mC'
    mosdepth_d4_output             = false
    filter_snvs_expression         = ''
    filter_svs_expression          = ''
    force_sawfish_joint_call_single_samples = false
    paraphrase_output_format       = 'tsv'
    paraphrase_skip_keys           = null
    paraphrase_genes               = null
    pre_vep_snv_filter_expression  = ''
    phaser                         = 'longphase'
    sv_callers                     = 'sniffles,hificnv'
    sv_callers_merge_priority      = params.sv_callers
    sv_callers_to_run              = params.sv_callers
    sv_callers_to_merge            = params.sv_callers
    run_methbat                    = params.preset == 'revio' ? true : false
    run_modkit                     = params.preset == 'ONT_R10' ? true : false
    hifiasm_mode                   = 'trio-binning'
    snv_caller                     = 'deepvariant'
    str_caller                     = params.preset == 'ONT_R10' ? 'strdust' : 'trgt'
    tiddit_bin_size                = 500
    vep_cache_version              = 110
    sentieon_model_bundle          = null
    sentieon_tech                  = params.preset == 'ONT_R10' ? 'ONT' : 'HiFi'
    sentieon_male_haploid_bed      = null
    sentieon_male_diploid_bed      = null
    sentieon_female_diploid_bed    = null
    strdrop_fraction               = 0.66
    strdrop_alpha                  = 0.18
    strdrop_edit                   = 0.9


    // Genmod compound scoring options
    genmod_compound_snv_penalty              = 6
    genmod_compound_snv_threshold            = 9
    genmod_compound_sv_penalty               = 6
    genmod_compound_sv_threshold             = 9
    genmod_compound_singleton_snv_penalty    = params.genmod_compound_snv_penalty
    genmod_compound_singleton_snv_threshold  = params.genmod_compound_snv_threshold
    genmod_compound_singleton_sv_penalty     = params.genmod_compound_sv_penalty
    genmod_compound_singleton_sv_threshold   = params.genmod_compound_sv_threshold
    genmod_compound_trio_snv_penalty         = params.genmod_compound_snv_penalty
    genmod_compound_trio_snv_threshold       = params.genmod_compound_snv_threshold
    genmod_compound_trio_sv_penalty          = params.genmod_compound_sv_penalty
    genmod_compound_trio_sv_threshold        = params.genmod_compound_sv_threshold

    // Test config options
    extra_hifiasm_options         = ''
    extra_methbat_profile_options = ''
    extra_modkit_options          = ''
    extra_paraphase_options       = ''
    extra_sawfish_options         = ''
    extra_vep_options             = ''
    extra_yak_options             = ''

    // Preprocessing/parallelization
    snv_calling_processes        = 13
    alignment_processes          = 8

    // 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                     = System.getenv('HOOK_URL')
    help                         = false
    help_full                    = false
    show_hidden                  = false
    version                      = false
    modules_testdata_base_path   = 'https://raw.githubusercontent.com/nf-core/test-datasets/modules/data/'
    pipelines_testdata_base_path = 'https://raw.githubusercontent.com/genomic-medicine-sweden/test-datasets/eaabae15e3ee2b074057363a7467469b09f0957e/'
    trace_report_suffix          = new java.util.Date().format( 'yyyy-MM-dd_HH-mm-ss')// 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

    // Schema validation default options
    validate_params            = true
}

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

profiles {
    debug {
        dumpHashes              = true
        process.beforeScript    = 'echo $HOSTNAME'
        cleanup                 = false
        nextflow.enable.configProcessNamesValidation = true
    }
    conda {
        conda.enabled           = true
        docker.enabled          = false
        singularity.enabled     = false
        podman.enabled          = false
        shifter.enabled         = false
        charliecloud.enabled    = false
        conda.channels          = ['conda-forge', 'bioconda']
        apptainer.enabled       = false
    }
    mamba {
        conda.enabled           = true
        conda.useMamba          = true
        docker.enabled          = false
        singularity.enabled     = false
        podman.enabled          = false
        shifter.enabled         = false
        charliecloud.enabled    = false
        apptainer.enabled       = false
    }
    docker {
        docker.enabled          = true
        conda.enabled           = false
        singularity.enabled     = false
        podman.enabled          = false
        shifter.enabled         = false
        charliecloud.enabled    = false
        apptainer.enabled       = false
        docker.runOptions       = '-u $(id -u):$(id -g)'
    }
    arm64 {
        process.arch            = 'arm64'
        // https://github.com/nf-core/modules/issues/6694
        // For now if you're using arm64 you have to use wave for the sake of the maintainers
        // wave profile
        apptainer.ociAutoPull   = true
        singularity.ociAutoPull = true
        wave.enabled            = true
        wave.freeze             = true
        wave.strategy           = 'conda,container'
    }
    emulate_amd64 {
        docker.runOptions       = '-u $(id -u):$(id -g) --platform=linux/amd64'
    }
    singularity {
        singularity.enabled     = true
        singularity.autoMounts  = true
        conda.enabled           = false
        docker.enabled          = false
        podman.enabled          = false
        shifter.enabled         = false
        charliecloud.enabled    = false
        apptainer.enabled       = false
    }
    podman {
        podman.enabled          = true
        conda.enabled           = false
        docker.enabled          = false
        singularity.enabled     = false
        shifter.enabled         = false
        charliecloud.enabled    = false
        apptainer.enabled       = false
    }
    shifter {
        shifter.enabled         = true
        conda.enabled           = false
        docker.enabled          = false
        singularity.enabled     = false
        podman.enabled          = false
        charliecloud.enabled    = false
        apptainer.enabled       = false
    }
    charliecloud {
        charliecloud.enabled    = true
        conda.enabled           = false
        docker.enabled          = false
        singularity.enabled     = false
        podman.enabled          = false
        shifter.enabled         = false
        apptainer.enabled       = false
    }
    apptainer {
        apptainer.enabled       = true
        apptainer.autoMounts    = true
        conda.enabled           = false
        docker.enabled          = false
        singularity.enabled     = false
        podman.enabled          = false
        shifter.enabled         = false
        charliecloud.enabled    = false
    }
    wave {
        apptainer.ociAutoPull   = true
        singularity.ociAutoPull = true
        wave.enabled            = true
        wave.freeze             = true
        wave.strategy           = 'conda,container'
    }
    gpu {
        docker.runOptions       = '-u $(id -u):$(id -g) --gpus all'
        apptainer.runOptions    = '--nv'
        singularity.runOptions  = '--nv'
    }
    test      { includeConfig 'tests/nextflow.config'      }
    test_full { includeConfig 'tests/nextflow.config'      }
}

// Load nf-core custom profiles from different institutions

// If params.custom_config_base is set AND either the NXF_OFFLINE environment variable is not set or params.custom_config_base is a local path, the nfcore_custom.config file from the specified base path is included.
// Load genomic-medicine-sweden/nallo custom profiles from different institutions.
includeConfig params.custom_config_base && (!System.getenv('NXF_OFFLINE') || !params.custom_config_base.startsWith('http')) ? "${params.custom_config_base}/nfcore_custom.config" : "/dev/null"


// Set default registry for Apptainer, Docker, Podman, Charliecloud and Singularity independent of -profile
// Will not be used unless Apptainer / Docker / Podman / Charliecloud / 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'
charliecloud.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"
}

// Set bash options
process.shell = [
    "bash",
    "-C",         // No clobber - prevent output redirection from overwriting files.
    "-e",         // Exit if a tool returns a non-zero status/exit code
    "-u",         // Treat unset variables and parameters as an error
    "-o",         // Returns the status of the last command to exit..
    "pipefail"    //   ..with a non-zero status or zero if all successfully execute
]

// Disable process selector warnings by default. Use debug profile to enable warnings.
nextflow.enable.configProcessNamesValidation = false

timeline {
    enabled = true
    file    = "${params.outdir}/pipeline_info/execution_timeline_${params.trace_report_suffix}.html"
}
report {
    enabled = true
    file    = "${params.outdir}/pipeline_info/execution_report_${params.trace_report_suffix}.html"
}
trace {
    enabled = true
    file    = "${params.outdir}/pipeline_info/execution_trace_${params.trace_report_suffix}.txt"
}
dag {
    enabled = true
    file    = "${params.outdir}/pipeline_info/pipeline_dag_${params.trace_report_suffix}.html"
}

manifest {
    name            = 'genomic-medicine-sweden/nallo'
    contributors    = [
        [
            name: 'Ramprasad Neethiraj',
            affiliation: 'School of Engineering sciences in Chemistry, Biotechnology and Health, KTH Royal Institute of Technology, Stockholm, Sweden; Science for Life Laboratory, Department of Microbiology, Tumour and Cell Biology, Karolinska Institutet, Stockholm, Sweden',
            email: 'rne@kth.se',
            github: 'https://github.com/ramprasadn',
            contribution: ['contributor'],
            orcid: 'https://orcid.org/0000-0001-7313-3734'
        ],
        [
            name: 'Anders Jemt',
            affiliation: 'Genomics Medicine Center Karolinska, Karolinska University Hospital, Stockholm, Sweden; Science for Life Laboratory, Department of Microbiology, Tumour and Cell Biology, Karolinska Institutet, Stockholm, Sweden',
            email: 'anders.jemt@scilifelab.se',
            github: 'https://github.com/jemten',
            contribution: ['contributor'],
            orcid: 'https://orcid.org/0000-0002-2219-0197'
        ],
        [
            name: 'Annick Renevey',
            affiliation: 'Science for Life Laboratory, Department of Microbiology, Tumour and Cell Biology, Karolinska Institutet, Stockholm, Sweden',
            email: 'annick.renevey@scilifelab.se',
            github: 'https://github.com/rannick',
            contribution: ['contributor'],
            orcid: 'https://orcid.org/0000-0001-7411-6063'
        ],
        [
            name: 'Daniel Schmitz',
            affiliation: 'Clinical Genomics Gothenburg, Science for Life Laboratory, Sahlgrenska Academy, University of Gothenburg.',
            email: 'daniel.schmitz@gu.se',
            github: 'https://github.com/Schmytzi',
            contribution: ['contributor'],
            orcid: 'https://orcid.org/0000-0003-4480-891X'
        ],
        [
            name: 'Felix Lenner',
            affiliation: 'Department of Immunology, Genetics and Pathology, Uppsala University, Sweden; Science for Life Laboratory, Department of Microbiology, Tumour and Cell Biology, Karolinska Institutet, Stockholm, Sweden',
            email: 'felix.lenner@scilifelab.se',
            github: 'https://github.com/fellen31',
            contribution: ['author','maintainer'],
            orcid: 'https://orcid.org/0000-0002-9594-0710'
        ],
        [
            name: 'Ine Marie Esbensen',
            affiliation: 'Science for Life Laboratory, Department of Microbiology, Tumour and Cell Biology, Karolinska Institutet, Stockholm, Sweden',
            email: 'ine-marie.esbensen@scilifelab.se',
            github: 'https://github.com/inemesb',
            contribution: ['contributor']
        ],
        [
            name: 'Jakob Willforss',
            affiliation: 'Section for Molecular Diagnostics, Region Skåne, Lund, Sweden',
            email: 'jakob.willforss@skane.se',
            github: 'https://github.com/Jakob37',
            contribution: ['contributor'],
            orcid: 'https://orcid.org/0000-0001-8183-8769'
        ],
        [
            name: 'Lucía Peña-Pérez',
            affiliation: 'Center for Inherited Metabolic Diseases, Karolinska University Hospital, Stockholm, Sweden; Department of Molecular Medicine and Surgery, Karolinska Institutet, Stockholm, Sweden',
            email: 'lucia.pena.perez@scilifelab.se',
            github: 'https://github.com/Lucpen',
            contribution: ['contributor'],
            orcid: 'https://orcid.org/0000-0002-5044-7754'
        ],
        [
            name: 'Padraic Corcoran',
            affiliation: 'Department of Immunology, Genetics and Pathology, Uppsala University, Uppsala, Sweden. Clinical genomics, Science for life laboratory, Uppsala University.',
            email: 'padraic.corcoran@scilifelab.uu.se',
            github: 'https://github.com/padraicc',
            contribution: ['contributor'],
        ],
        [
            name: 'Peter Pruisscher',
            affiliation: 'Genomics Medicine Center Karolinska, Karolinska University Hospital, Stockholm, Sweden; Science for Life Laboratory, Department of Microbiology, Tumour and Cell Biology, Karolinska Institutet, Stockholm, Sweden',
            email: 'peter.pruisscher@scilifelab.se',
            github: 'https://github.com/peterpru',
            contribution: ['contributor'],
            orcid: 'https://orcid.org/0000-0002-6987-5839'
        ],
        [
            name: 'Alexander Koc',
            affiliation: 'Medical Services, Skåne University Hospital, Lund, Sweden',
            email: 'alexander.koc@skane.se',
            github: 'https://github.com/alkc',
            contribution: ['contributor'],
        ],
    ]
    homePage        = 'https://github.com/genomic-medicine-sweden/nallo'
    description     = """Long-read variant calling pipeline"""
    mainScript      = 'main.nf'
    defaultBranch   = 'master'
    nextflowVersion = '!>=25.10.0'
    version         = '0.12.0dev'
    doi             = ''
}

// Nextflow plugins
plugins {
    id 'nf-schema@2.6.1' // Validation of pipeline parameters and creation of an input channel from a sample sheet
}

validation {
    defaultIgnoreParams = ["genomes"]
    monochromeLogs = params.monochrome_logs
}

// Load modules.config for DSL2 module specific options
includeConfig 'conf/modules/align_assemblies.config'
includeConfig 'conf/modules/annotate_cadd.config'
includeConfig 'conf/modules/annotate_consequence_pli.config'
includeConfig 'conf/modules/annotate_paralogs.config'
includeConfig 'conf/modules/annotate_repeat_expansions.config'
includeConfig 'conf/modules/annotate_snvs.config'
includeConfig 'conf/modules/annotate_svs.config'
includeConfig 'conf/modules/bam_infer_sex.config'
includeConfig 'conf/modules/call_methylation_modkit.config'
includeConfig 'conf/modules/call_methylation_methbat.config'
includeConfig 'conf/modules/call_paralogs.config'
includeConfig 'conf/modules/call_repeat_expansions_trgt.config'
includeConfig 'conf/modules/call_repeat_expansions_strdust.config'
includeConfig 'conf/modules/call_snvs.config'
includeConfig 'conf/modules/call_svs.config'
includeConfig 'conf/modules/chromograph.config'
includeConfig 'conf/modules/convert_input_files.config'
includeConfig 'conf/modules/filter_variants.config'
includeConfig 'conf/modules/genome_assembly.config'
includeConfig 'conf/modules/gvcf_glnexus_norm_variants.config'
includeConfig 'conf/modules/hiphase.config'
includeConfig 'conf/modules/longphase.config'
includeConfig 'conf/modules/nallo.config'
includeConfig 'conf/modules/phasing.config'
includeConfig 'conf/modules/prepare_references.config'
includeConfig 'conf/modules/qc_aligned_reads.config'
includeConfig 'conf/modules/qc_phasing.config'
includeConfig 'conf/modules/qc_snvs.config'
includeConfig 'conf/modules/rank_variants.config'
includeConfig 'conf/modules/scatter_genome.config'
includeConfig 'conf/modules/split_multisample_vcf.config'
includeConfig 'conf/modules/vcf_concat_norm_variants.config'
includeConfig 'conf/modules/vcf_concat_sort_variants.config'
includeConfig 'conf/modules/whatshap.config'
includeConfig 'conf/modules/prepare_gens_input.config'