nextflow.config

/*
Pipeline parameters
*/
// Glob to capture paired-end reads, should capture R1 and R2 with brace expansion
params.reads = 'data/fastq_subset/SRR1258338*_R{1,2}.fastq.gz'
// Final output dir
params.outdir = 'results'
// If wanting to use Kraken with a DB that is already on file, use this param:
params.krakendb_dir = null
// If wanting to use Kraken with a DB that can be downloaded from the internet, use this param
// (see https://benlangmead.github.io/aws-indexes/k2. Example: 'https://genome-idx.s3.amazonaws.com/kraken/k2_pluspf_20240605.tar.gz')
params.krakendb_url = 'https://genome-idx.s3.amazonaws.com/kraken/k2_pluspfp_20240605.tar.gz'
// If wanting to use Kraken with a custom-built DB, use this param to provide a dir with genomes to add:  
params.krakendb_add = null
// If wanting to use Kraken with a custom-built DB, use this param to provide a comma-separated list of 1 or more libraries
// The library options are: 'archaea', 'bacteria', 'viral', 'plasmid', 'human', 'fungi', 'plant', 'protozoa', 'UniVec', 'Univec_Core', 'nr', 'nt'
params.krakendb_libs = 'plant'

// Software environments
singularity {
    enabled = true
    autoMounts = true
    cacheDir = '/fs/scratch/PAS0471/jelmer/containers'
}

// Process-specific config
process {
    errorStrategy = 'finish' // change to 'retry' in final version

    withName: 'KRAKEN_RUN' {
        memory = { 175.GB * task.attempt }
        cpus = 12
    }
    withName: 'KRAKEN_BUILD' {
        time = 12.h
        memory = { 100.GB * task.attempt }
        cpus = 12
    }
    withName: 'KRAKEN_DL_LIB' {
        time = 12.h
        memory = 4.GB
        cpus = 1
    }
    withName: 'ASSEMBLY' {
        memory = 40.GB
        cpus = 10
    }
}

// Profiles
profiles {
    standard {
        process.executor = 'slurm'
        process.clusterOptions = '--account=PAS2693'
        process.time = { 60.min * task.attempt }  // Or: '5 min'
        process.cpus = { 2 * task.attempt }
        process.memory = { 4.GB * task.cpus }  // Or: '8 GB' , 5.MB
        process.resourceLimits = [ time: 168.h ]
    }

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

    test_slurm {
        process.executor = 'slurm'
        process.clusterOptions = '-A PAS1568'
        process.memory = 4.GB
        process.cpus = 1
        process.time = 1.h
        process.resourceLimits = [ cpus: 4, memory: 16.GB, time: 10.h ]
    }
}