4 implement conversion from spaceranger output to zarr and h5ad

.nf-core.yml

@@ -31,7 +31,7 @@ repository_type: pipeline
 template:
   author: M. Bonfanti, S. Terzoli
   description: Analysis of spatial omics dataset
-  force: true 
+  force: true
   is_nfcore: false
   name: spatialomics
   org: nfdata-omics

CHANGELOG.md

@@ -9,8 +9,13 @@ Initial release of nfdata-omics/spatialomics, created with the [nf-core](https:/
 
 ### `Added`
 
+- Implemented alignment with spaceranger [2](https://github.com/nfdata-omics/spatialomics/pull/2)
+- Implemented conversion from spaceranger output to zarr, possibility of direct input of a spaceranger out folder, collecting metrics file from spaceranger to multiQC [7](https://github.com/nfdata-omics/spatialomics/pull/7)
+
 ### `Fixed`
 
+- Template update to 3.5.2 [3](https://github.com/nfdata-omics/spatialomics/pull/3)
+
 ### `Dependencies`
 
 ### `Deprecated`

assets/multiqc_config.yml

@@ -2,6 +2,10 @@ report_comment: >
   This report has been generated by the <a href="https://github.com/nfdata-omics/spatialomics/tree/dev" target="_blank">nfdata-omics/spatialomics</a>
   analysis pipeline.
 report_section_order:
+  fastqc:
+    order: 5000
+  spaceranger_metrics:
+    order: 4000
   "nfdata-omics-spatialomics-methods-description":
     order: -1000
   software_versions:
@@ -12,3 +16,29 @@ report_section_order:
 export_plots: true
 
 disable_version_detection: true
+
+# Run only these modules
+run_modules:
+  - custom_content
+  - fastqc
+  - spaceranger
+
+# Customize file name patterns
+sp:
+  spaceranger:
+    fn: "*web_summary.html"
+    contents: None
+  spaceranger_metrics:
+    fn: "spaceranger_metrics.csv"
+
+custom_data:
+  spaceranger_metrics:
+    id: "spaceranger_metrics"
+    section_name: "Space Ranger metrics"
+    description: |
+      This table summarizes key metrics from the Space Ranger analysis across all samples.
+      It includes information such as the number of reads, the percentage of reads mapped to the probe set,
+      and other relevant statistics that can help assess the quality of the spatial transcriptomics data.
+    plot_type: "table"
+    pconfig:
+      id: "spaceranger_metrics_table"

assets/schema_input.json

@@ -68,8 +68,14 @@
                 "format": "file-path",
                 "exists": true,
                 "meta": ["slidefile"]
+            },
+            "spaceranger": {
+                "type": "string",
+                "format": "path",
+                "exists": true,
+                "errorMessage": "Path to Space Ranger output directory must exist"
             }
         },
-        "required": ["sample", "fastq_1"]
+        "anyOf": [{ "required": ["sample", "fastq_1"] }, { "required": ["sample", "spaceranger"] }]
     }
 }

conf/modules.config

@@ -102,20 +102,41 @@ process {
             [
                 path: { "${params.outdir}/count/" },
                 mode: params.publish_dir_mode,
-                pattern: "outs/web_summary.html",
+                pattern: "*web_summary.html",
                 saveAs: { "${meta.id}_web_summary.html" }
             ],
             [
-                path: { "${params.outdir}/count/${meta.id}" },
+                path: { "${params.outdir}/count/" },
                 mode: params.publish_dir_mode,
-                pattern: "outs/**",
-                saveAs: { filename -> filename.contains('/') ? filename.substring(filename.indexOf('/') + 1) : filename }
+                pattern: "outs",
+                saveAs: { "${meta.id}" }
             ]
         ]
         ext.args   = "--create-bam true"
         time   = { 240.h * task.attempt }
     }
 
+    withName: 'COLLECT_SPACERANGER_METRICS' {
+        publishDir = [
+            enabled: false
+        ]
+    }
+
+    withName: 'SPACERANGER_TO_ZARR' {
+        publishDir = [
+            enabled: false
+        ]
+    }
+
+    withName: 'TAR' {
+        ext.prefix = { "${meta.id}.zarr" }
+        publishDir = [
+            path: "${params.outdir}/count",
+            mode: params.publish_dir_mode,
+            saveAs: { filename -> filename.equals('versions.yml') ? null : filename }
+        ]
+    }
+
     withName: 'MULTIQC' {
         ext.args   = { params.multiqc_title ? "--title \"$params.multiqc_title\"" : '' }
         publishDir = [

main.nf

@@ -41,6 +41,7 @@ workflow NFDATAOMICS_SPATIALOMICS {
 
     take:
     samplesheet // channel: samplesheet read in from --input
+    spaceranger_outs // channel: spaceranger output paths read in from --input
 
     main:
 
@@ -56,6 +57,7 @@ workflow NFDATAOMICS_SPATIALOMICS {
     //
     SPATIALOMICS (
         samplesheet,
+        spaceranger_outs,
         ch_fasta,
         ch_gtf,
         ch_gff,
@@ -93,7 +95,8 @@ workflow {
     // WORKFLOW: Run main workflow
     //
     NFDATAOMICS_SPATIALOMICS (
-        PIPELINE_INITIALISATION.out.samplesheet
+        PIPELINE_INITIALISATION.out.samplesheet,
+        PIPELINE_INITIALISATION.out.spaceranger_outs
     )
 
     //

modules.json

@@ -51,6 +51,11 @@
                         "git_sha": "41dfa3f7c0ffabb96a6a813fe321c6d1cc5b6e46",
                         "installed_by": ["modules"]
                     },
+                    "tar": {
+                        "branch": "master",
+                        "git_sha": "5c9f8d5b7671237c906abadc9ff732b301ca15ca",
+                        "installed_by": ["modules"]
+                    },
                     "untar": {
                         "branch": "master",
                         "git_sha": "00ee87ebb541af0008596400ce6d5f66d79d5408",

modules/local/collect_spaceranger_metrics/main.nf

@@ -0,0 +1,101 @@
+process COLLECT_SPACERANGER_METRICS {
+    tag "all samples"
+    label 'process_low'
+
+    container 'docker.io/nfdata/spatialdata:v0.7.2'
+
+    input:
+    path "outs_*"
+
+    output:
+    path "spaceranger_metrics.csv",  emit: metrics
+    path "versions.yml", emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    """
+    #!/usr/bin/env python3
+
+    import sys
+    import importlib
+    import pkg_resources
+    import glob
+    import yaml
+    import pandas as pd
+
+    input_files = glob.glob("outs_*/metrics_summary.csv")
+
+    dfs = []
+    for file in input_files:
+        df = pd.read_csv(file)
+        dfs.append(df)
+
+    # Concatenate all DataFrames
+    combined = pd.concat(dfs, ignore_index=True)
+
+    # Save to a single CSV file
+    combined.to_csv("spaceranger_metrics.csv", index=False)
+
+    # ----------------------------------
+    # Print versions of relevant libraries
+    # ----------------------------------
+
+    versions = {}
+    versions["${task.process}"] = {}
+    for lib in ['pandas']:
+        try:
+            version = pkg_resources.get_distribution(lib).version
+        except Exception:
+            try:
+                module = importlib.import_module(lib)
+                version = getattr(module, '__version__', 'unknown')
+            except Exception:
+                version = None
+        if version is not None:
+            versions["${task.process}"][lib] = version
+    versions["${task.process}"]['python'] = f"{sys.version_info.major}.{sys.version_info.minor}.{sys.version_info.micro}"
+
+    with open('versions.yml', 'w') as f:
+        yaml.dump(versions, f)
+    """
+
+    stub:
+    """
+    #!/usr/bin/env python3
+
+    import sys
+    import importlib
+    import pkg_resources
+    import glob
+    import yaml
+
+    open('spaceranger_metrics.csv', 'w').close()
+
+    # ----------------------------------
+    # Print versions of relevant libraries
+    # ----------------------------------
+
+    versions = {}
+    versions["${task.process}"] = {}
+    for lib in ['pandas']:
+        try:
+            version = pkg_resources.get_distribution(lib).version
+        except Exception:
+            try:
+                module = importlib.import_module(lib)
+                version = getattr(module, '__version__', 'unknown')
+            except Exception:
+                version = None
+        if version is not None:
+            versions["${task.process}"][lib] = version
+    versions["${task.process}"]['python'] = f"{sys.version_info.major}.{sys.version_info.minor}.{sys.version_info.micro}"
+
+    with open('versions.yml', 'w') as f:
+        yaml.dump(versions, f)
+
+    """
+
+
+}

modules/local/spaceranger_to_zarr/main.nf

@@ -0,0 +1,106 @@
+process SPACERANGER_TO_ZARR {
+    tag "$meta.id"
+    label 'process_low'
+
+    container 'docker.io/nfdata/spatialdata:v0.7.2'
+
+    input:
+    tuple val(meta), path(spaceranger_output_dir)
+    val filtered_counts_file
+
+    output:
+    tuple val(meta), path("*.zarr"), emit: zarr
+    path "versions.yml"            , emit: versions
+
+    when:
+    task.ext.when == null || task.ext.when
+
+    script:
+    def prefix = task.ext.prefix ?: "${meta.id}"
+    """
+    #!/usr/bin/env python3
+
+    import os
+
+    os.environ["NUMBA_CACHE_DIR"] = os.environ.get("TMPDIR", "/tmp")
+    os.environ["MPLCONFIGDIR"] = os.environ.get("TMPDIR", "/tmp")
+    os.environ["XDG_CONFIG_HOME"] = os.environ.get("TMPDIR", "/tmp")
+
+    import sys
+    import importlib
+    import pkg_resources
+    import yaml
+    import spatialdata_io
+
+    # ----------------------------------
+    # Load dataset using spatialdata_io
+    # ----------------------------------
+
+    # Here, the bin size is set to 16, but we should allow the option to also read 002 and 008
+    data = spatialdata_io.visium_hd(
+        "$spaceranger_output_dir",
+        filtered_counts_file="${filtered_counts_file}",
+        dataset_id="${prefix}"
+    )
+
+    # ----------------------------------
+    # Save full dataset as Zarr
+    # ----------------------------------
+    data.write(f"${prefix}.zarr", overwrite=True)
+
+    # ----------------------------------
+    # Print versions of relevant libraries
+    # ----------------------------------
+
+    versions = {}
+    versions["${task.process}"] = {}
+    for lib in ['spatialdata_io', 'spatialdata', 'numpy', 'pandas', 'scipy']:
+        try:
+            version = pkg_resources.get_distribution(lib).version
+        except Exception:
+            try:
+                module = importlib.import_module(lib)
+                version = getattr(module, '__version__', 'unknown')
+            except Exception:
+                version = None
+        if version is not None:
+            versions["${task.process}"][lib] = version
+    versions["${task.process}"]['python'] = f"{sys.version_info.major}.{sys.version_info.minor}.{sys.version_info.micro}"
+
+    with open('versions.yml', 'w') as f:
+        yaml.dump(versions, f)
+    """
+
+    stub:
+    """
+    #!/usr/bin/env python3
+
+    import os
+    os.makedirs("${meta.id}.zarr", exist_ok=True)
+
+    # ----------------------------------
+    # Print versions of relevant libraries
+    # ----------------------------------
+
+    versions = {}
+    versions["${task.process}"] = {}
+    for lib in ['spatialdata_io', 'spatialdata', 'numpy', 'pandas', 'scipy']:
+        try:
+            version = pkg_resources.get_distribution(lib).version
+        except Exception:
+            try:
+                module = importlib.import_module(lib)
+                version = getattr(module, '__version__', 'unknown')
+            except Exception:
+                version = None
+        if version is not None:
+            versions["${task.process}"][lib] = version
+    versions["${task.process}"]['python'] = f"{sys.version_info.major}.{sys.version_info.minor}.{sys.version_info.micro}"
+
+    with open('versions.yml', 'w') as f:
+        yaml.dump(versions, f)
+
+    """
+
+
+}