nextflow.config

// Default parameters
params {

    // Input runs as a TSV of `cohort`, `path`, and `type`
    input_tsv = null

    // global output directories - should be changed for each deployment
    // the paths in this repository are really just for the test workflow
    processed_annotations = "nextflow/processed_annotations"

    // outdir's default location is for running the processed_annotations workflow
    // designed to be overridden for real workflows (checked on startup)
    outdir = "${params.processed_annotations}"

    // Large files directory
    large_files = "large_files"
    ref_genome = "${params.large_files}/ref.fa"

    // Docker container - "docker build -f docker/Dockerfile -t talos:12.0.3 ."
    container = 'talos:12.0.3'

    // GATK Docker container, for use in optional SV annotation pathway. Pulled from public source, no local build
    gatk_container = 'broadinstitute/gatk:4.6.2.0'

    // SVAFotate container - "docker build -f docker/SVAFotate_Dockerfile -t svafotate:0.1.0 ."
    // built separately from the main Talos image, its pinned dependencies are irreconcilable with Talos's
    svafotate_container = 'svafotate:0.1.0'

    // MANE transcript resource
    mane = "${params.large_files}/MANE.GRCh38.v1.5.summary.txt.gz"

    // Result file from running ParseManeIntoJson on the MANE summary file
    mane_json = "${params.processed_annotations}/mane.json"

    // Ensembl gene feature file
    ensembl_gff = "${params.large_files}/Homo_sapiens.GRCh38.116.MTtoM.gff3.gz"

    // Result files from the CreateRoiFromGff3 nextflow module
    ensembl_bed = "${params.processed_annotations}/GRCh38.bed"
    ensembl_merged_bed = "${params.processed_annotations}/GRCh38_merged.bed"
    ensembl_symbol_lookup = "${params.processed_annotations}/GRCh38_symbol_to_ensg.json"

    // AlphaMissense raw data, and reformatted as an echtvar-compatible zip
    alphamissense_tsv = "${params.large_files}/AlphaMissense_hg38.tsv.gz"
    alphamissense_zip = "${params.processed_annotations}/alphamissense.zip"

    // A series of annotation sources for Mitochondrial data
    mitimpact_tsv = "${params.large_files}/MitImpact_db_3.1.3.txt.zip"
    mitimpact_zip = "${params.processed_annotations}/mitimpact.zip"

    mitotip_tsv = "${params.large_files}/mitotip_scores.txt"
    mitotip_zip = "${params.processed_annotations}/mitotip.zip"

    napogee_tsv = "${params.large_files}/nAPOGEE_v1.0.0.txt.zip"
    napogee_zip = "${params.processed_annotations}/napogee.zip"

    // ClinVar download URLs
    submission_summary = "https://ftp.ncbi.nlm.nih.gov/pub/clinvar/tab_delimited/submission_summary.txt.gz"
    variant_summary = "https://ftp.ncbi.nlm.nih.gov/pub/clinvar/tab_delimited/variant_summary.txt.gz"

    // ClinVar blacklist
    clinvar_blacklist = "'not' 'applicable'"

    // HPO and Phenotype resources
    hpo = "${params.large_files}/hp.obo"
    gen2phen = "${params.large_files}/genes_to_phenotype.txt"
    phenio_db = "${params.large_files}/phenio.db"

    // ------------------------------------------------------------------------
    // Annotation Workflow Parameters
    // ------------------------------------------------------------------------

    // Input VCF extension and location logic
    input_vcf_extension = "vcf.bgz"
    vcf_split_n = 2500000

    // Echtvar-formatted annotation data
    gnomad_zip = "${params.large_files}/gnomad_4.1_region_merged_GRCh38_whole_genome"

    // ------------------------------------------------------------------------
    // SV Annotation Workflow Parameters
    // ------------------------------------------------------------------------

    // The joint-called SV VCF to annotate is supplied per-cohort, as the `sv` column of the input TSV

    // SVAFotate population frequency BED, gnomAD v4.1 + CCDG + TOPMed + 1000G
    // used exactly as downloaded - the Ensembl-style contig names in this file are required, see docs
    svafotate_bed = "${params.large_files}/SVAFotate_reduced_gnomAD.bed.gz"

    // MANE GTF, consumed by GATK SVAnnotate. One transcript per gene, as SVAnnotate requires
    mane_gtf = "${params.large_files}/MANE.GRCh38.v1.5.ensembl_genomic.gtf.gz"

    // Non-coding elements BED from the GATK-SV public resources, consumed by GATK SVAnnotate.
    // Adds PREDICTED_NONCODING_BREAKPOINT and PREDICTED_NONCODING_SPAN - informational, Talos reads neither yet
    svannotate_noncoding_bed = "${params.large_files}/noncoding.sort.hg38.bed"

    // Sequence dictionary generated from ref_genome, required by GATK SVAnnotate for contig ordering
    ref_dict = "${params.processed_annotations}/ref.dict"

    // Reciprocal overlap threshold for matching SVs against the gnomAD reference
    sv_overlap_fraction = 0.5
}

// switch with -with-docker or -without-docker cli parameters
docker.enabled = true

// set the -output-dir via config (seqera doesn't currently allow this to be set via CLI)
outputDir = params.outdir

workflow.output.mode = 'copy'

// Process configuration
process {
    memory = 8.GB
    cpus = 2

    // Specific process resource requirements
    withName: 'ResummariseRawSubmissions' {
        memory = 16.GB
        cpus = 4
    }
    withName: 'SplitVcf' {
        memory = 2.GB
        cpus = 1
    }
    withName: 'AnnotateCsqWithBcftools' {
        memory = 1.GB
        cpus = 1
    }
    withName: 'AnnotateWithEchtvar' {
        cpus = 1
        memory = 1.GB
    }
    withName: 'AnnotatedVcfIntoMatrixTable' {
        memory = 8.GB
        cpus = 4
    }
    withName: 'RunHailFiltering' {
        memory = 8.GB
        cpus = 4
    }
    withName: 'NormaliseVcf' {
        memory = 1.GB
        cpus = 1
    }
    withName: 'CreateRoiFromGff3' {
        memory = 2.GB
        cpus = 1
    }
    withName: 'DownloadClinVarFiles' {
        memory = 2.GB
    }
    withName: 'EncodeAlphaMissense' {
        memory = 16.GB
    }
    withName: 'CreateSequenceDictionary' {
        memory = 4.GB
        cpus = 1
    }
    withName: 'AnnotateSvWithGatk' {
        memory = 4.GB
        cpus = 1
    }
    withName: 'AnnotateSvWithSvafotate' {
        memory = 16.GB
        cpus = 4
    }
    withName: 'RenameSvAfFields' {
        memory = 2.GB
        cpus = 1
    }
    withName: 'ValidateMOI' {
        memory = 16.GB
    }
}