Multisample support for spatialxe runs

assets/multiqc_config.yml

@@ -14,25 +14,19 @@ export_plots: true
 
 disable_version_detection: true
 
-run_module:
-  - xenium
-
-module_order:
-  - xenium
-
 log_filesize_limit: 5000000000 # 5GB
 
 sp:
-  cell_feature_matrix:
-    fn: cell_feature_matrix.h5
-  cells:
-    fn: cells.parquet
-  experiment:
-    fn: experiment.xenium
+  xenium/cell_feature_matrix:
+    fn: "*/cell_feature_matrix.h5"
+  xenium/cells:
+    fn: "*/cells.parquet"
+  xenium/experiment:
+    fn: "*/experiment.xenium"
     num_lines: 50
-  metrics:
+  xenium/metrics:
     contents: num_cells_detected
-    fn: metrics_summary.csv
+    fn: "*/metrics_summary.csv"
     num_lines: 5
-  transcripts:
-    fn: transcripts.parquet
+  xenium/transcripts:
+    fn: "*/transcripts.parquet"

conf/modules.config

@@ -32,13 +32,15 @@ process {
             path: "${params.outdir}/xeniumranger/resegment",
             mode: params.publish_dir_mode
         ]
+        ext.prefix = {"${meta.id}"}
     }
 
     withName: XENIUMRANGER_IMPORT_SEGMENTATION {
         publishDir = [
             path: "${params.outdir}/xeniumranger/import_segementation",
             mode: params.publish_dir_mode
         ]
+        ext.prefix = {"${meta.id}"}
     }
 
     withName: FICTURE_PREPROCESS {
@@ -152,4 +154,24 @@ process {
         ext.args = { "--diameter 9 --channel_axis 0 --save_flows" }
     }
 
+    withName: OPT_FLIP {
+        publishDir = [
+            path: { "${params.outdir}/opt/flip" },
+            mode: params.publish_dir_mode,
+        ]
+    }
+
+    withName: OPT_TRACK {
+        publishDir = [
+            path: { "${params.outdir}/opt/track" },
+            mode: params.publish_dir_mode,
+        ]
+    }
+
+    withName: OPT_STAT {
+        publishDir = [
+            path: { "${params.outdir}/opt/stat" },
+            mode: params.publish_dir_mode,
+        ]
+    }
 }

modules.json

@@ -19,17 +19,20 @@
                     "opt/flip": {
                         "branch": "master",
                         "git_sha": "66d5baa4e9b6ac3ab95e84f88709b8e3ebf4b62b",
-                        "installed_by": ["modules"]
+                        "installed_by": ["modules"],
+                        "patch": "modules/nf-core/opt/flip/opt-flip.diff"
                     },
                     "opt/stat": {
                         "branch": "master",
                         "git_sha": "66d5baa4e9b6ac3ab95e84f88709b8e3ebf4b62b",
-                        "installed_by": ["modules"]
+                        "installed_by": ["modules"],
+                        "patch": "modules/nf-core/opt/stat/opt-stat.diff"
                     },
                     "opt/track": {
                         "branch": "master",
                         "git_sha": "66d5baa4e9b6ac3ab95e84f88709b8e3ebf4b62b",
-                        "installed_by": ["modules"]
+                        "installed_by": ["modules"],
+                        "patch": "modules/nf-core/opt/track/opt-track.diff"
                     },
                     "untar": {
                         "branch": "master",

modules/local/baysor/create_dataset/main.nf

@@ -9,8 +9,8 @@ process BAYSOR_CREATE_DATASET {
     val(sample_fraction)
 
     output:
-    tuple val(meta), path("sampled_transcripts.csv"), emit: sampled_transcripts
-    path("versions.yml")                            , emit: versions
+    tuple val(meta), path("${prefix}/sampled_transcripts.csv"), emit: sampled_transcripts
+    path("versions.yml")                                      , emit: versions
 
     when:
     task.ext.when == null || task.ext.when
@@ -21,6 +21,8 @@ process BAYSOR_CREATE_DATASET {
         error "BAYSOR_CREATE_DATASET module does not support Conda. Please use Docker / Singularity / Podman instead."
     }
 
+    prefix = task.ext.prefix ?: "${meta.id}"
+
     template 'create_dataset.py'
 
     stub:
@@ -29,8 +31,11 @@ process BAYSOR_CREATE_DATASET {
         error "BAYSOR_CREATE_DATASET module does not support Conda. Please use Docker / Singularity / Podman instead."
     }
 
+    prefix = task.ext.prefix ?: "${meta.id}"
+
     """
-    touch sampled_transcripts.csv
+    mkdir -p ${prefix}
+    touch "${prefix}/sampled_transcripts.csv"
 
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":

modules/local/baysor/create_dataset/templates/create_dataset.py

@@ -14,7 +14,8 @@ def generate_dataset(
             transcripts: Path,
             sampled_transcripts: Path,
             sample_fraction: float = 0.3,
-            random_state: int = 42
+            random_state: int = 42,
+            prefix: str = ""
         ) -> None:
         """
         Reads a csv file & randomly samples a fraction of rows,
@@ -28,8 +29,8 @@ def generate_dataset(
         """
 
         random.seed(random_state)
-
-        with open(transcripts, mode='rt', newline='') as infile, \
+        output_path = f"{prefix}/{transcripts}"
+        with open(output_path, mode='rt', newline='') as infile, \
             open(sampled_transcripts, mode='wt', newline='') as outfile:
 
             reader = csv.reader(infile)
@@ -61,13 +62,15 @@ def main() -> None:
     """
     transcripts: str = "${transcripts}"
     sample_fraction: float = "${sample_fraction}"
+    prefix: str = "${meta.id}"
     sampled_transcripts: str = "sampled_transcripts.csv"
 
     # generate dataset
     BaysorPreview.generate_dataset (
         transcripts=transcripts,
         sampled_transcripts=sampled_transcripts,
-        sample_fraction=sample_fraction
+        sample_fraction=sample_fraction,
+        prefix=prefix
     )
 
     # generate versions.yml

modules/local/baysor/preprocess/main.nf

@@ -13,22 +13,35 @@ process BAYSOR_PREPROCESS_TRANSCRIPTS {
     val(min_y)
 
     output:
-    tuple val(meta), path("*.parquet"), emit: transcripts_parquet
-    path("versions.yml")              , emit: versions
+    tuple val(meta),
+          path("${prefix}/filtered_transcripts.parquet"), emit: transcripts_parquet
+    path("versions.yml")                                , emit: versions
 
     when:
     task.ext.when == null || task.ext.when
 
     script:
+    // Exit if running this module with -profile conda / -profile mamba
     if (workflow.profile.tokenize(',').intersect(['conda', 'mamba']).size() >= 1) {
-        error "PREPROCESS_TRANSCRIPTS module does not support Conda. Please use Docker / Singularity / Podman instead."
+        error "BAYSOR_PREPROCESS_TRANSCRIPTS module does not support Conda. Please use Docker / Singularity / Podman instead."
     }
 
+    prefix = task.ext.prefix ?: "${meta.id}"
+
     template 'preprocess_transcripts.py'
 
     stub:
+    // Exit if running this module with -profile conda / -profile mamba
+    if (workflow.profile.tokenize(',').intersect(['conda', 'mamba']).size() >= 1) {
+        error "BAYSOR_PREPROCESS_TRANSCRIPTS module does not support Conda. Please use Docker / Singularity / Podman instead."
+    }
+
+    prefix = task.ext.prefix ?: "${meta.id}"
+
     """
-    touch ${transcripts}.parquet
+    mkdir -p ${prefix}
+    touch ${prefix}/filtered_transcripts.parquet
+
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":
         baysor_preprocess_transcripts: "1.0.0"

modules/local/baysor/preprocess/templates/preprocess_transcripts.py

@@ -9,7 +9,8 @@ def filter_transcripts (
     min_x: float = 0.0,
     max_x: float = 24000.0,
     min_y: float = 0.0,
-    max_y: float = 24000.0
+    max_y: float = 24000.0,
+    prefix: str = ""
 ) -> None:
     """
     Filter transcripts based on the specified thresholds
@@ -48,7 +49,7 @@ def filter_transcripts (
 
     # Output filtered transcripts to parquet
     filtered_df.to_parquet(
-        '_'.join(["X"+str(min_x)+"-"+str(max_x), "Y"+str(min_y)+"-"+str(max_y), "filtered_transcripts.parquet"]),
+        '_'.join(["X"+str(min_x)+"-"+str(max_x), "Y"+str(min_y)+"-"+str(max_y), f"{prefix}/filtered_transcripts.parquet"]),
         index=False
     )
 
@@ -66,9 +67,11 @@ def generate_version_yml() -> None:
 if __name__ == "__main__":
 
     transcripts: str = "${transcripts}"
+    prefix: str = "${meta.id}"
 
     filter_transcripts (
         transcripts=transcripts,
+        prefix=prefix
     )
 
     generate_version_yml()

modules/local/baysor/preview/main.nf

@@ -9,9 +9,8 @@ process BAYSOR_PREVIEW {
     path(config)
 
     output:
-    tuple val(meta), path("preview.html"), emit: preview_html
-    path("preview_preview_log.log")      , emit: preview_log
-    path("versions.yml")                 , emit: versions
+    tuple val(meta), path("${prefix}/preview.html"), emit: preview_html
+    path("versions.yml")                           , emit: versions
 
     when:
     task.ext.when == null || task.ext.when
@@ -21,14 +20,20 @@ process BAYSOR_PREVIEW {
     if (workflow.profile.tokenize(',').intersect(['conda', 'mamba']).size() >= 1) {
         error "BAYSOR_PREVIEW module does not support Conda. Please use Docker / Singularity / Podman instead."
     }
+
     def args = task.ext.args ?: ''
+    prefix = task.ext.prefix ?: "${meta.id}"
 
     """
+    mkdir -p ${prefix}
+
     baysor preview \\
     ${transcripts} \\
     --config ${config} \\
     ${args}
 
+    mv preview.html ${prefix}/preview.html
+
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":
         baysor: 0.7.1
@@ -41,9 +46,11 @@ process BAYSOR_PREVIEW {
         error "BAYSOR_PREVIEW module does not support Conda. Please use Docker / Singularity / Podman instead."
     }
 
+    prefix = task.ext.prefix ?: "${meta.id}"
+
     """
-    touch preview.html
-    touch preview_preview_log.log
+    mkdir -p ${prefix}
+    touch "${prefix}/preview.html"
 
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":

modules/local/baysor/run/main.nf

@@ -5,21 +5,17 @@ process BAYSOR_RUN {
     container "khersameesh24/baysor:0.7.1"
 
     input:
-    tuple val(meta), path(transcripts)
-    path(prior_segmentation)
-    path(config)
-    val(scale)
+    tuple val(meta),
+          path(transcripts),
+          path(prior_segmentation),
+          path(config),
+          val(scale)
 
     output:
-    tuple val(meta), path("segmentation.csv"), emit: segmentation
-    path("segmentation_polygons_2d.json")    , emit: polygons2d
-    path("segmentation_polygons_3d.json")    , emit: polygons3d
-    path("*.toml")                           , emit: params
-    path("*.log")                            , emit: log
-    path("*.loom")                           , emit: loom
-    path("*.html")                           , emit: htmls
-    path("segmentation_cell_stats.csv")      , emit: stats
-    path("versions.yml")                     , emit: versions
+    tuple val(meta),
+          path("${prefix}/segmentation.csv"),
+          path("${prefix}/segmentation_polygons_2d.json"), emit: segmentation
+    path("versions.yml")                                 , emit: versions
 
     when:
     task.ext.when == null || task.ext.when
@@ -29,15 +25,20 @@ process BAYSOR_RUN {
     if (workflow.profile.tokenize(',').intersect(['conda', 'mamba']).size() >= 1) {
         error "BAYSOR_RUN module does not support Conda. Please use Docker / Singularity / Podman instead."
     }
+
     def args = task.ext.args ?: ''
     def prior_seg = "${prior_segmentation}" ? "${prior_segmentation}" : ""
     def scaling_factor = scale ? "--scale=${scale}": ""
+    prefix = task.ext.prefix ?: "${meta.id}"
 
     """
+    mkdir -p ${prefix}
+
     baysor run \\
     ${transcripts} \\
     ${prior_seg} \\
     ${scaling_factor} \\
+    --output="${prefix}/segmentation.csv" \\
     --config=${config} \\
     --plot \\
     --polygon-format=GeometryCollectionLegacy \\
@@ -55,15 +56,12 @@ process BAYSOR_RUN {
         error "BAYSOR_RUN module does not support Conda. Please use Docker / Singularity / Podman instead."
     }
 
+    prefix = task.ext.prefix ?: "${meta.id}"
+
     """
-    touch segmentation.csv
-    touch segmentation_polygons_2d.json
-    touch segmentation_polygons_3d.json
-    touch segmentation_log.log
-    touch segmentation_counts.loom
-    touch segmentation_cell_stats.csv
-    touch segmentation_params.dump.toml
-    touch segmentation_run.html
+    mkdir -p ${prefix}
+    touch "${prefix}/segmentation.csv"
+    touch "${prefix}/segmentation_polygons_2d.json"
 
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":