nextflow.config

/*
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    nf-core/spatialaxe Nextflow config file
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    Default config options for all compute environments
----------------------------------------------------------------------------------------
*/

// Global default params, used in configs
params {

    // Input options
    input                      = null       // path to the samplesheet.csv containing meta,bundle,image
    outdir                     = null       // path to generate pipeline results at
    mode                       = null       // please check nextflow_schema.json for the modes you can select
    method                     = null       // name of the method to run for image or coordinate or segfree approaches
    gene_panel                 = null       // path to gene panel json file if `relabel_genes` is true
    qupath_polygons            = null       // Path to qupath segmentation results in GeoJSON format
    alignment_csv              = null       // image alignment file format a 3x3 transformation matrix, where the last row is [0,0,1]
    cellpose_model             = null       // custom cellpose model to use for running or starting training
    stardist_model             = '2D_versatile_fluo'  // stardist pretrained model for cell segmentation
    stardist_nuclei_model      = '2D_versatile_fluo'  // stardist pretrained model for nuclei segmentation
    stardist_prob_thresh       = null       // stardist probability threshold
    stardist_nms_thresh        = null       // stardist NMS threshold
    stardist_n_tiles           = "8 8"      // tiling for large images (Xenium images are ~20K×25K)
    segmentation_mask          = null       // prior segmentation mask
    probes_fasta               = null       // Fasta file for the probe sequences used in the xenium experiment
    reference_annotations      = null       // Path to the genomic features (.gff) and fasta (.fa) files used as reference annotations
    gene_synonyms              = null       // Gene synonyms that may have been counted as off-targets but simply differ in name

    // GPU flag (set to true by the gpu profile)
    use_gpu                    = false
    gpu_queue                  = null       // AWS Batch queue for GPU tasks (e.g., SEGGER, ProSeg)
    cellpose_queue             = null       // AWS Batch queue for Cellpose (single large GPU)

    // execution specific
    sharpen_tiff               = false      // wether to sharpen the morphology-focus tiff
    nucleus_segmentation_only  = false      // to only run nucleus segmentation while running segmentation methods & XR_IMP-SEG
    cell_segmentation_only     = true       // to only run cell segmentation while running segmentation methods & XR_IMP-SEG
    cellpose_downscale         = false      // pre-downscale morphology image to avoid cellpose OOM on large images

    // Xeniumranger specific
    xeniumranger_only          = false      // to generate redefined bundle with just changing the xr specific params
    relabel_genes              = false      // wether to correct gene names with gene_panel.json
    expansion_distance         = 5          // default nuclear expansion distance in XOA v2.0 & later
    dapi_filter                = 100        // adjust the minimum peak intensity to use more nuclei
    interior_stain             = true       // interior stain is enabled by default - false to disable
    boundary_stain             = true       // boundary stain is enabled by default - false to disable

    // Segger specific
    segmentation_refinement    = false      // wether to run segmentation refinement step (segger)
    segger_accelerator         = 'cpu'      // either 'cuda' or 'cpu'
    segger_knn_method          = 'kd_tree'  // 'cuda' - ensure your system has CUDA installed and configured properly
    segger_num_workers         = 4          // number of data-loader workers for segger
    segger_model               = null       // path to a pre-trained segger model checkpoint

    // Proseg specific
    format                     = 'xenium'   // preset value set as `xenium`

    // Segmentation methods
    image_seg_methods          = ["cellpose", "xeniumranger", "baysor", "stardist"]
    transcript_seg_methods     = ["proseg", "segger", "baysor"]
    segfree_methods            = ["ficture", "baysor"]

    // Ficture specific
    negative_control_regex     = null
    features                   = null

    // Baysor specific
    filter_transcripts         = false
    min_qv                     = 20
    max_x                      = 24000.0
    min_x                      = 0.0
    max_y                      = 24000.0
    min_y                      = 0.0

    // Generic tiling parameters (for proseg and other methods)
    tiling                      = false          // enable tiled segmentation (divide → parallel segmentation → stitch)
    patch_grid                  = '3x3'          // grid layout for tiling (rows x cols)
    patch_overlap               = 50             // overlap between patches in microns
    patch_filter_method         = null           // post-stitch cell filtering: 'empirical', 'distribution', 'both', or null
    patch_filter_iqr_multiplier = 3.0            // IQR multiplier for empirical cell size filtering
    patch_filter_z_threshold    = 4.0            // z-score threshold for distribution cell size filtering

    // Baysor-specific parameters
    baysor_scale                = 30             // Baysor --scale for non-tiled runs
    baysor_config               = null           // path to baysor config TOML (optional)
    baysor_tiling               = true           // enable tiled Baysor (divide → per-patch Baysor → stitch)
    baysor_tiling_micron        = 1200           // tile width in microns for Baysor tiling
    baysor_tiling_overlap       = 200            // overlap between Baysor patches in microns
    baysor_tiling_balanced      = true           // balance transcripts across tiles (merge sparse tiles)
    baysor_tiling_scale         = 39             // Baysor --scale for tiled runs (larger to compensate for EM on tiles)
    baysor_tiling_min_mols_per_cell    = 120     // --min-molecules-per-cell for tiled Baysor
    baysor_tiling_min_transcripts_per_cell = 50  // post-stitch cell filtering threshold

    // Baysor prior segmentation
    //   null       — no prior (random EM init)
    //   'cells'    — Xenium bundle cell_id column (column-based, works with tiling)
    //   'cellpose' — run Cellpose cell mask as image prior (non-tiled only)
    baysor_prior                = null
    baysor_prior_confidence     = 0.2            // prior-segmentation-confidence [0-1]

    // Segger specific
    tile_width                 = 120
    tile_height                = 120
    batch_size_train           = 4          // larger batch size can speed up training, but requires more memory
    devices                    = 4          // Use multiple GPUs by increasing the devices parameter to further accelerate training
    max_epochs                 = 200        // increasing #epochs can improve model performance with more learning cycles, but extends training time
    batch_size_predict         = 1          // larger batch size can speed up training, but requires more memory
    cc_analysis                = false      // to control connected component analysis

    // qc specific
    run_qc                     = true       // whether to run the qc layer of pipeline
    offtarget_probe_tracking   = false      // whether to run off-target probe tracking (provide probe_fasta, reference sequences, gene synonyms )

    // utility modules
    csplit_x_bins              = 2          // number of tiles along the x axis (total number of bins is product of x_bins * y_bins)
    csplit_y_bins              = 2          // number of tiles along the y axis

    // MultiQC options
    multiqc_config             = null
    multiqc_title              = null
    multiqc_logo               = null
    max_multiqc_email_size     = '25.MB'
    multiqc_methods_description = null

    // pipeline dev and testing option
    buffer_samples             = false       // process one sample at a time from the multi-sample samplesheet
    buffer_size                = 1           // buffer size 0 means no buffering of samples

    // Boilerplate options
    publish_dir_mode             = 'copy'
    email                        = null
    email_on_fail                = null
    plaintext_email              = false
    monochromeLogs               = false
    monochrome_logs              = false
    hook_url                     = System.getenv('HOOK_URL')
    help                         = false
    help_full                    = false
    show_hidden                  = false
    version                      = false
    pipelines_testdata_base_path = 'https://raw.githubusercontent.com/nf-core/test-datasets/'
    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
    test_data_base             = 'https://raw.githubusercontent.com/nf-core/test-datasets/spatialaxe'


    // 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           = false
        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           = false
        conda.useMamba          = false
        docker.enabled          = false
        singularity.enabled     = false
        podman.enabled          = false
        shifter.enabled         = false
        charliecloud.enabled    = false
        apptainer.enabled       = false
    }
    docker {
        docker.enabled         = true
        docker.fixOwnership    = 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)'
    }
    arm {
        docker.runOptions      = '-u $(id -u):$(id -g) --platform=linux/amd64'
    }
    singularity {
        singularity.enabled     = true
        singularity.autoMounts  = true
        singularity.pullTimeout = '100 min'
        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'
    }
    gitpod {
        executor.name           = 'local'
        executor.cpus           = 8
        executor.memory         = 16.GB
    }
    gpu {
        params.use_gpu          = true
        docker.runOptions       = '-u $(id -u):$(id -g) --gpus all'
        apptainer.runOptions    = '--nv'
        singularity.runOptions  = '--nv'
    }
    aws {
        process {
            withLabel:process_gpu {
                // Must repeat base.config label properties — profile withLabel replaces, not merges
                ext.use_gpu = { params.use_gpu }
                accelerator = { params.use_gpu ? 1 : null }
                containerOptions = { "--shm-size ${task.memory.toGiga()}g" }
                queue = { params.gpu_queue ?: null }
            }
            withLabel:process_gpu_single {
                ext.use_gpu = { params.use_gpu }
                accelerator = { params.use_gpu ? 1 : null }
                containerOptions = { "--shm-size ${task.memory.toGiga()}g" }
                queue = { params.cellpose_queue ?: params.gpu_queue ?: null }
            }
        }
    }
    test      { includeConfig 'conf/test.config'      }
    test_full { includeConfig 'conf/test_full.config' }
}


// Load nf-core 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"

// Load nf-core/spatialaxe custom profiles from different institutions.
includeConfig !System.getenv('NXF_OFFLINE') && params.custom_config_base ? "${params.custom_config_base}/pipeline/spatialaxe.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 {
    PYTORCH_CUDA_ALLOC_CONF = "expandable_segments:True"
    PYTHONNOUSERSITE = 1
    R_PROFILE_USER   = "/.Rprofile"
    R_ENVIRON_USER   = "/.Renviron"
    JULIA_DEPOT_PATH = "/usr/local/share/julia"
    MPLCONFIGDIR     = "./tmp"
    NUMBA_CACHE_DIR      = "./tmp"
    NUMBA_DISABLE_CACHE  = 1
}

// 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
nextflow.enable.moduleBinaries       = true

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
    overwrite = 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            = 'nf-core/spatialaxe'
    contributors    = [
        [
            name: 'Sameesh Kher',
            affiliation: 'German Cancer Research Center (DKFZ), Heidelberg, DE',
            email: 'sameesh.kher@dkfz-heidelberg.de',
            github: '@khersameesh24',
            contribution: ['author', 'maintainer'],
            orcid: '0009-0008-2420-6464'
        ],
        [
            name: 'Florian Heyl',
            affiliation: 'German Cancer Research Center (DKFZ), Heidelberg, DE',
            email: 'florian.heyl@dkfz-heidelberg.de',
            github: '@heylf',
            contribution: ['author', 'maintainer'],
            orcid: ''
        ],
        [
            name: 'Dongze He',
            affiliation: 'Altos Labs, San Diego, USA',
            email: 'dongzehe.zaza@gmail.com',
            github: '@dongzehe',
            contribution: ['contributor'],
            orcid: '0000-0001-8259-7434'
        ]
    ]
    homePage        = 'https://github.com/nf-core/spatialaxe'
    description     = """A pipeline for spatialomics 10x Xenium In Situ data."""
    mainScript      = 'main.nf'
    defaultBranch   = 'master'
    nextflowVersion = '!>=25.04.0'
    version         = '1.0.1'
    doi             = ''
}

// Nextflow plugins
plugins {
    id 'nf-schema@2.5.1' // 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.config'