Initiate module for cell-type-ewings

.pre-commit-config.yaml

@@ -37,5 +37,5 @@ repos:
       - id: style-files
       - id: parsable-R
 ci:
-  autofix_prs: true
+  autofix_prs: false
   autoupdate_schedule: quarterly

config/containers.config

@@ -20,4 +20,8 @@ params{
 
   // cell-type-consensus module
   consensus_cell_type_container = 'public.ecr.aws/openscpca/cell-type-consensus:v0.2.2'
+
+  // cell-type-ewings module
+  cell_type_ewing_container = 'public.ecr.aws/openscpca/cell-type-ewings:v0.2.2'
+
 }

config/module_params.config

@@ -13,4 +13,9 @@ params{
   cell_type_consensus_ref_file = 'https://raw.githubusercontent.com/AlexsLemonade/OpenScPCA-analysis/refs/tags/v0.2.2/analyses/cell-type-consensus/references/consensus-cell-type-reference.tsv'
   cell_type_consensus_marker_gene_ref_file = 'https://raw.githubusercontent.com/AlexsLemonade/OpenScPCA-analysis/refs/tags/v0.2.2/analyses/cell-type-consensus/references/validation-markers.tsv'
 
+  // cell type ewings
+  cell_type_ewings_auc_max_rank = 425 // 1% of the total detected genes in the merged object, equivalent to the value used to obtain assignments in OpenScPCA-analysis
+  cell_type_ewings_msigdb_list = 'https://raw.githubusercontent.com/AlexsLemonade/OpenScPCA-analysis/refs/tags/v0.2.2/analyses/cell-type-ewings/references/msigdb-gene-sets.tsv'
+  cell_type_ewings_ews_high_list = 'https://raw.githubusercontent.com/AlexsLemonade/OpenScPCA-analysis/refs/tags/v0.2.2/analyses/cell-type-ewings/references/gene_signatures/aynaud-ews-targets.tsv'
+  cell_type_ewings_ews_low_list = 'https://raw.githubusercontent.com/AlexsLemonade/OpenScPCA-analysis/refs/tags/v0.2.2/analyses/cell-type-ewings/references/gene_signatures/wrenn-nt5e-genes.tsv'
 }

main.nf

@@ -7,6 +7,7 @@ include { merge_sce } from './modules/merge-sce'
 include { detect_doublets } from './modules/doublet-detection'
 include { seurat_conversion } from './modules/seurat-conversion'
 include { cell_type_consensus } from './modules/cell-type-consensus'
+include { cell_type_ewings } from './modules/cell-type-ewings'
 
 // **** Parameter checks ****
 include { validateParameters; paramsSummaryLog } from 'plugin/nf-schema'
@@ -52,14 +53,17 @@ workflow {
     .filter{ run_all || it[1] in project_ids }
 
   // Run the merge workflow
-  merge_sce(sample_ch)
+  //merge_sce(sample_ch)
 
   // Run the doublet detection workflow
-  detect_doublets(sample_ch)
+  //detect_doublets(sample_ch)
 
   // Run the seurat conversion workflow
-  seurat_conversion(sample_ch)
+  //seurat_conversion(sample_ch)
 
   // Run the consensus cell type workflow
-  cell_type_consensus(sample_ch)
+  //cell_type_consensus(sample_ch)
+
+  // Run the cell type ewings workflow
+  cell_type_ewings(sample_ch)
 }

modules/cell-type-ewings/README.md

@@ -0,0 +1,13 @@
+This module assigns cell types to all Ewing sarcoma samples in `SCPCP000015`.
+
+Scripts are derived from the the `cell-type-ewings` module of the [OpenScPCA-analysis](https://github.com/AlexsLemonade/OpenScPCA-analysis) repository.
+
+Links to specific original scripts used in this module:
+
+- `01-aucell.R`: <https://github.com/AlexsLemonade/OpenScPCA-analysis/blob/main/analyses/cell-type-ewings/scripts/aucell-ews-signatures/01-aucell.R>
+
+This module also uses the following reference files found in the `OpenScPCA-analysis` repository:
+
+- `aynaud-ews-targets.tsv` : <https://github.com/AlexsLemonade/OpenScPCA-analysis/refs/tags/v0.2.1/analyses/cell-type-ewings/references/gene_signatures/aynaud-ews-targets.tsv>
+- `wrenn-nt5e-genes.tsv`: <https://github.com/AlexsLemonade/OpenScPCA-analysis/refs/tags/v0.2.1/analyses/cell-type-ewings/references/gene_signatures/wrenn-nt5e-genes.tsv>
+- `msigdb-gene-sets.tsv`: <https://github.com/AlexsLemonade/OpenScPCA-analysis/refs/tags/v0.2.1/analyses/cell-type-ewings/references/msigdb-gene-sets.tsv>

modules/cell-type-ewings/main.nf

@@ -0,0 +1,80 @@
+#!/usr/bin/env nextflow
+
+// Workflow to assign consensus cell type labels
+
+process ewing_aucell {
+  container params.cell_type_ewing_container
+  tag "${project_id}"
+  label 'mem_8'
+  publishDir "${params.results_bucket}/${params.release_prefix}/cell-type-ewings/${project_id}/${sample_id}", mode: 'copy'
+  input:
+    tuple val(sample_id),
+          val(project_id),
+          path(library_files)
+    val auc_max_rank
+    path msigdb_list
+    path ews_high_list
+    path ews_low_list
+  output:
+    tuple val(sample_id),
+          val(project_id),
+          path(aucell_output_files)
+  script:
+    aucell_output_files = library_files
+      .collect{
+        it.name.replaceAll(/(?i).rds$/, "_ewing-aucell-results.tsv.gz")
+      }
+    // combine the custom gene sets into a single input
+    custom_geneset_files = [ews_high_list, ews_low_list].join(",")
+    """
+    for file in ${library_files}; do
+      aucell.R \
+        --sce_file \$file \
+        --custom_geneset_files ${custom_geneset_files} \
+        --msigdb_genesets ${msigdb_list} \
+        --max_rank_threshold ${auc_max_rank} \
+        --output_file \$(basename \${file%.rds}_ewing-aucell-results.tsv.gz) \
+        --threads ${task.cpus} \
+        --seed 2025
+    done
+    """
+
+  stub:
+    aucell_output_files = library_files
+      .collect{
+        it.name.replaceAll(/(?i).rds$/, "_ewing-aucell-results.tsv.gz")
+      }
+    """
+    for file in ${library_files}; do
+      touch \$(basename \${file%.rds}_ewing-aucell-results.tsv.gz)
+    done
+    """
+}
+
+
+
+workflow cell_type_ewings {
+  take:
+    sample_ch  // [sample_id, project_id, sample_path]
+  main:
+    // create [sample_id, project_id, [list of processed files]]
+    libraries_ch = sample_ch
+      .map{sample_id, project_id, sample_path ->
+        def library_files = Utils.getLibraryFiles(sample_path, format: "sce", process_level: "processed")
+        return [sample_id, project_id, library_files]
+      }
+      // only run on SCPCP000015 with Ewing sarcoma samples
+      .filter{ it[1] == "SCPCP000015" }
+
+    // run aucell on ewing gene sets
+    ewing_aucell(
+      libraries_ch,
+      params.cell_type_ewings_auc_max_rank,
+      file(params.cell_type_ewings_msigdb_list),
+      file(params.cell_type_ewings_ews_high_list),
+      file(params.cell_type_ewings_ews_low_list)
+    )
+
+  emit:
+    ewing_aucell.out // [sample_id, project_id, [list of aucell_output_files]]
+}