Implement symphony integration and reference mapping

.gitignore

@@ -9,3 +9,6 @@ testing*
 null/
 .lineage/
 .nf-test*
+tests/assets/*.h5ad
+contrib/nf-core-test-datasets/build_output/
+contrib/nf-core-test-datasets/extension_base/

README.md

@@ -58,7 +58,7 @@ Steps marked with the boat icon are not yet implemented. For the other steps, th
    3. Integration
       - [scVI](https://docs.scvi-tools.org/en/stable/user_guide/models/scvi.html)
       - [scANVI](https://docs.scvi-tools.org/en/stable/user_guide/models/scanvi.html)
-      - [Harmony](https://portals.broadinstitute.org/harmony/articles/quickstart.html)
+      - [Symphony](https://github.com/immunogenomics/symphony) / Harmony (via [symphonypy](https://pypi.org/project/symphonypy/))
       - [BBKNN](https://github.com/Teichlab/bbknn)
       - [Combat](https://scanpy.readthedocs.io/en/latest/api/generated/scanpy.pp.combat.html)
       - [Seurat](https://satijalab.org/seurat/articles/integration_introduction)

assets/multiqc_config.yml

@@ -18,12 +18,12 @@ report_section_order:
     order: -1005
   "scanvi":
     order: -1006
-  "harmony":
+  "symphony":
     order: -1007
   "bbknn":
-    order: -1008
-  "combat":
     order: -1009
+  "combat":
+    order: -1010
   # If new tools are add. They need to be added here
   "nf-core-scdownstream-methods-description":
     order: -2001

assets/schema_analysis_plan.json

@@ -12,27 +12,31 @@
                 "pattern": "^\\S*$",
                 "default": null,
                 "errorMessage": "Integration name cannot contain spaces",
-                "description": "Integration method name, or empty to match all integrations"
+                "description": "Integration method name, or empty to match all integrations",
+                "meta": ["integration"]
             },
             "subset": {
                 "type": "string",
                 "pattern": "^\\S*$",
                 "default": null,
                 "errorMessage": "Subset cannot contain spaces",
-                "description": "Clustering subset (global or a label value), or empty to match all subsets"
+                "description": "Clustering subset (global or a label value), or empty to match all subsets",
+                "meta": ["subset"]
             },
             "resolution": {
                 "type": "number",
                 "minimum": 0,
                 "default": null,
-                "description": "Leiden resolution, or empty to match all resolutions"
+                "description": "Leiden resolution, or empty to match all resolutions",
+                "meta": ["resolution"]
             },
             "analyses": {
                 "type": "string",
                 "pattern": "^(|paga|liana|de|cytetype)(,(paga|liana|de|cytetype))*$",
                 "default": null,
                 "errorMessage": "Analyses must be a comma-separated list of paga, liana, de, and/or cytetype",
-                "description": "Downstream analyses to run for matching clusterings, or empty to run all analyses"
+                "description": "Downstream analyses to run for matching clusterings, or empty to run all analyses",
+                "meta": ["analyses"]
             }
         }
     }

conf/modules.config

@@ -379,9 +379,25 @@ process {
         ]
     }
 
-    withName: SCANPY_HARMONY {
+    withName: SYMPHONY_HARMONYINTEGRATE {
         publishDir = [
-            path: { "${params.outdir}/combine/integrate/harmony" },
+            path: { "${params.outdir}/combine/integrate/symphony" },
+            mode: params.publish_dir_mode,
+            saveAs: { filename ->
+                if (filename.endsWith('_reference.h5ad')) {
+                    return 'symphony_reference.h5ad'
+                }
+                if (params.save_intermediates && !filename.equals('versions.yml')) {
+                    return filename
+                }
+                return null
+            },
+        ]
+    }
+
+    withName: SYMPHONY_MAPEMBEDDING {
+        publishDir = [
+            path: { "${params.outdir}/combine/integrate/symphony" },
             mode: params.publish_dir_mode,
             enabled: params.save_intermediates,
             saveAs: { filename -> filename.equals('versions.yml') ? null : filename },

conf/test.config

@@ -24,7 +24,7 @@ params {
 
     // Input data
     input               = params.pipelines_testdata_base_path + 'samplesheet.csv'
-    integration_methods = 'scvi,harmony,bbknn,combat'
+    integration_methods = 'scvi,symphony,bbknn,combat'
     doublet_detection   = 'solo,scrublet,scdblfinder'
     celltypist_model    = 'Adult_Human_Skin'
     celldex_reference   = 'https://raw.githubusercontent.com/nf-core/test-datasets/scdownstream/singleR/references.csv'

conf/test_full.config

@@ -24,7 +24,7 @@ params {
 
     // Input data for full size test
     input               = params.pipelines_testdata_base_path + 'samplesheet.csv'
-    integration_methods = 'scvi,harmony,bbknn,combat'
+    integration_methods = 'scvi,symphony,bbknn,combat'
     doublet_detection   = 'solo,scrublet,doubletdetection,scdblfinder'
     celltypist_model    = 'Adult_Human_Skin'
     celldex_reference   = 'https://raw.githubusercontent.com/nf-core/test-datasets/scdownstream/singleR/references.csv'

docs/output.md

@@ -34,7 +34,7 @@ The pipeline is built using [Nextflow](https://www.nextflow.io/) and processes d
    3. Integration
       - [scVI](https://docs.scvi-tools.org/en/stable/user_guide/models/scvi.html)
       - [scANVI](https://docs.scvi-tools.org/en/stable/user_guide/models/scanvi.html)
-      - [Harmony](https://portals.broadinstitute.org/harmony/articles/quickstart.html)
+      - [Symphony](https://github.com/immunogenomics/symphony) / Harmony (via [symphonypy](https://pypi.org/project/symphonypy/))
       - [BBKNN](https://github.com/Teichlab/bbknn)
       - [Combat](https://scanpy.readthedocs.io/en/latest/api/generated/scanpy.pp.combat.html)
       - [Seurat](https://satijalab.org/seurat/articles/integration_introduction)
@@ -98,6 +98,7 @@ The `preprocess` directory contains a subdirectory for each sample, which contai
     - `${tool}`
       - `*.h5ad/*.rds`: The integrated H5AD or RDS file.
       - `X_${tool}.pkl`: Low-dimensional representation of the integrated data.
+      - `symphony_reference.h5ad` (Symphony only): Compact Symphony reference AnnData for query mapping, published from de novo Symphony runs.
 
 </details>