params {
    config_profile_description = 'UW Hyak Roberts labs cluster profile provided by nf-core/configs.'
    config_profile_contact = 'Steven Roberts'
    config_profile_url = 'https://faculty.washington.edu/sr320/'

    /*
     * Desired downsample size per FASTQ (paired => each file limited to this many READS, i.e. 100k lines /4).
     * NOTE: nf-core/methylseq does NOT have a native param for subsetting; create 100k-read FASTQs prior to run
     * (see usage notes at bottom of file). This value is here for documentation and potential templating.
     */
    subset_reads = 100000

    /*
     * List of Bismark --score_min settings to evaluate. The pipeline itself executes a single value per run.
     * Loop over these externally (see usage notes). Typical form: L,0,-X  where X controls stringency.
     */
    score_min_values = [ 'L,0,-0.2', 'L,0,-0.4', 'L,0,-0.6', 'L,0,-0.8' ]

    /*
     * Single score_min value actually applied for this run. Override at runtime:
     *   --score_min 'L,0,-0.4'
     * Defaults to the most stringent (-0.8) in the list above.
     */
    score_min = 'L,0,-0.8'
}
 
process {
    errorStrategy = 'retry'
    maxSubmitAwait = '60 min'
    maxRetries = 2
    executor = 'slurm'
    queue  = { task.attempt < 4 ? (task.attempt < 3 ? 'ckpt-all' : 'cpu-g2' ) : 'cpu-g2-mem2x' }
    clusterOptions = { task.attempt < 4 ? (task.attempt < 3 ? "-A srlab" : "-A coenv" ) : "-A srlab" }
    // Use user-specific scratch on gscratch to avoid writing large temps to $HOME
    scratch = "/mmfs1/gscratch/scrubbed/${System.getenv('USER')}/nf-scratch"
    resourceLimits = [
        cpus: 48,
        memory: '400.GB',
        time: '72.h'
    ]

    withName: preseq { 
        errorStrategy = 'ignore' 
    }

    withName: BISMARK_ALIGN {
        cpus    = 32
        memory  = '100.GB'
        time    = '48.h'
        /*
         * We pass a single --score_min derived from params.score_min. Run the pipeline multiple times with different
         * values to compare (see bottom usage notes). Example override: --score_min 'L,0,-0.4'
         */
        ext.args = [
            "--score_min ${params.score_min}".toString(),
        ].join(' ')
        ext.prefix = { "${meta.id}_scmin_${params.score_min.replaceAll('[^0-9]+','')}" }
    }

    withName: BISMARK_METHYLATIONEXTRACTOR {
        ext.args = [
            '--merge_non_CpG',
            '--comprehensive',
            '--multicore 24',
            '--buffer_size 75%'
        ].join(' ')
        ext.prefix = { "${meta.id}" }
    }

    withName: BISMARK_COVERAGE2CYTOSINE {
        ext.args = [
            '--merge_CpG',
            '--zero_based'
        ].join(' ')
        ext.prefix = { "${meta.id}" }
    }
}

 
executor {
    queuesize = 50
    submitRateLimit = '1 sec'
}
 
singularity {
    enabled = true
    autoMounts = true
    // Redirect Apptainer/Singularity image cache off $HOME
    cacheDir = "/mmfs1/gscratch/scrubbed/${System.getenv('USER')}/singularity-cache"
}

trace {
    enabled = true
    file = 'pipeline_trace.txt'
    fields = 'task_id,hash,name,status,exit,submit,duration,realtime,%cpu,rss,vmem,rchar,wchar,disk,queue,hostname,cpu_model'
}
 
debug {
    cleanup = false
}

/* -------------------------------------------------------------------------
USAGE NOTES FOR SCORE_MIN SWEEP + READ SUBSETTING
-----------------------------------------------------------------------------
1. Create 100k-read subsets (per mate) once (paired example using seqtk):
         for fq in data/*_R1.fastq.gz; do \
             base=$(basename "$fq" _R1.fastq.gz); \
             seqtk sample -s100 $fq 100000 | gzip > data/subset/${base}_R1_100k.fastq.gz; \
             seqtk sample -s100 data/${base}_R2.fastq.gz 100000 | gzip > data/subset/${base}_R2_100k.fastq.gz; \
         done
     (Make sure to mkdir -p data/subset first.)

2. Prepare a samplesheet pointing to the subset FASTQs (e.g. 05.samplesheet.csv) and run one pipeline execution per score:
         for s in L,0,-0.2 L,0,-0.4 L,0,-0.6 L,0,-0.8; do \
             nextflow run nf-core/methylseq \
                 -profile <your_profiles> \
                 -c code/05.config \
                 --input code/05.samplesheet.csv \
                 --outdir output/05-scoremin-${s//[,.-]/} \
                 --score_min "$s"; \
         done

3. Results will be separated by outdir; prefixes include the numeric portion of score_min for clarity.

4. Compare alignment rates and methylation metrics across score_min thresholds (MultiQC, Bismark reports).

This block is purely informational; it does not alter pipeline execution.
----------------------------------------------------------------------------*/