nextflow.config

//
// nf-seabed-symphony — main configuration
//

manifest {
    name            = 'nf-seabed-symphony'
    description     = 'Nextflow implementation of the seabed-symphony metagenomics BGCs identification pipeline'
    author          = 'Data Science Platform - BRIGHT'
    homePage        = 'https://github.com/biosustain/dsp_nf_seabed_symphony'
    mainScript      = 'main.nf'
    version         = '0.1.0'
    // Topic channels + eval outputs + process.resourceLimits require >= 24.10
    nextflowVersion = '!>=24.10.0'
}

// ── Parameters ────────────────────────────────────────────────────────────────
params {
    // Input / output
    input       = null
    outdir      = 'results'

    // Workflow 2 — assembly
    skip_assembly    = false
    // Bandage is a visualisation side-branch. Skip it when running -profile
    // conda on macOS: the nf-core module uses GNU `zcat`, and the BSD `zcat`
    // that macOS ships refuses .gz files. Works fine in containers / on Linux.
    skip_bandage     = false
    // Flye read type, WITHOUT the leading dashes (Nextflow would otherwise read
    // a '--nano-hq' value as a new parameter). One of:
    //   nano-hq | nano-raw | nano-corr | pacbio-hifi | pacbio-raw | pacbio-corr
    // nano-hq needs <5 % read error (R10.4 / Q20+); use nano-raw for older R9.
    flye_read_type   = 'nano-hq'
    flye_iterations  = 3            // polishing rounds (original used 3)

    // Resource ceilings for the machine or queue you are running on
    max_cpus    = 16
    max_memory  = '128.GB'
    max_time    = '240.h'
}

// ── Resource ceilings ─────────────────────────────────────────────────────────
// Native Nextflow capping: any request from conf/base.config is clamped to these.
process.resourceLimits = [
    cpus  : params.max_cpus,
    memory: params.max_memory,
    time  : params.max_time
]

// ── Sub-configs ───────────────────────────────────────────────────────────────
includeConfig 'conf/base.config'
includeConfig 'conf/modules.config'

// ── Profiles ──────────────────────────────────────────────────────────────────
profiles {

    conda {
        conda.enabled       = true
        conda.useMamba      = false
        docker.enabled      = false
        singularity.enabled = false
    }

    mamba {
        conda.enabled       = true
        conda.useMamba      = true
        docker.enabled      = false
        singularity.enabled = false
    }

    docker {
        docker.enabled      = true
        docker.runOptions   = '-u $(id -u):$(id -g)'
        conda.enabled       = false
        singularity.enabled = false
    }

    singularity {
        singularity.enabled    = true
        singularity.autoMounts = true
        conda.enabled          = false
        docker.enabled         = false
    }

    // Use the locally built one-tool-per-image containers from docker/ instead
    // of the upstream images the nf-core modules declare. Build them first —
    // see docker/README.md.
    local_containers {
        docker.enabled      = true
        docker.runOptions   = '-u $(id -u):$(id -g)'
        conda.enabled       = false
        singularity.enabled = false

        process {
            withName: '.*:(NANOPLOT|NANOPLOT_TRIMMED)' {
                container = 'seabed-nanoplot:1.47.0'
            }
            withName: '.*:(SEQKIT_STATS|SEQKIT_STATS_TRIMMED)' {
                container = 'seabed-seqkit:2.13.0'
            }
            withName: '.*:(FILTLONG_MINLEN|FILTLONG_QUALITY)' {
                container = 'seabed-filtlong:0.2.1'
            }
            withName: '.*:PORECHOP_PORECHOP' {
                container = 'seabed-porechop:0.2.4'
            }
            withName: '.*:FLYE' {
                container = 'seabed-flye:2.9.5'
            }
            withName: '.*:BANDAGE_IMAGE' {
                container = 'seabed-bandage:0.9.0'
            }
            withName: '.*:WHOKARYOTE' {
                container = 'seabed-whokaryote:1.1.2'
            }
            withName: '.*:(EXTRACT_PROKARYOTE|EXTRACT_EUKARYOTE)' {
                container = 'seabed-biopython:1.84'
            }
        }
    }

    // Smoke test on a small, real bacterial ONT dataset (1 000 reads, ~33 kb
    // mean length, ~93 % pass the 1 kb filter). Streamed straight from the
    // nf-core test-datasets repo, so nothing large is stored in this repo.
    test {
        params.input  = 'https://raw.githubusercontent.com/nf-core/test-datasets/modules/data/genomics/prokaryotes/bacteroides_fragilis/nanopore/fastq/test.fastq.gz'
        params.outdir = 'test_results'

        process.executor       = 'local'
        process.resourceLimits = [ cpus: 4, memory: '8.GB', time: '1.h' ]
    }
}

// ── Execution reports ─────────────────────────────────────────────────────────
timeline { enabled = true; file = "${params.outdir}/pipeline_info/execution_timeline.html" ; overwrite = true }
report   { enabled = true; file = "${params.outdir}/pipeline_info/execution_report.html"   ; overwrite = true }
trace    { enabled = true; file = "${params.outdir}/pipeline_info/execution_trace.txt"     ; overwrite = true }
dag      { enabled = true; file = "${params.outdir}/pipeline_info/pipeline_dag.html"       ; overwrite = true }