generate test resources with wf

scripts/generate_test_resource.sh

@@ -1,16 +1,19 @@
 #!/bin/bash
 
-viash run src/process_datasets/convert/config.vsh.yaml -- \
+nextflow run . \
+  -main-script target/nextflow/workflows/process_datasets/main.nf \
   --input_sc resources_test/datasets_raw/MOBNEW/dataset_sc.rds \
   --input_sp resources_test/datasets_raw/MOBNEW/dataset_sp.rds \
   --output_sc resources_test/datasets/MOBNEW/dataset_sc.h5ad \
-  --output_sp resources_test/datasets/MOBNEW/dataset_sp.h5ad \
+  --output_sp resources_test/datasets/MOBNEW/temp_dataset_sp_part1.h5ad \
+  --id MOBNEW \
   --dataset_id MOBNEW \
   --dataset_name "MOBNEW" \
   --dataset_description "MOBNEW" \
   --dataset_summary "MOBNEW" \
   --dataset_reference "..." \
-  --dataset_organism "mus_musculus"
+  --dataset_organism "mus_musculus" \
+  --publish_dir resources_test/datasets
 
 viash run src/methods/scdesign3/config.vsh.yaml -- \
   --input resources_test/datasets/MOBNEW/dataset_sp.h5ad \

src/api/comp_process_datasets.yaml

@@ -27,12 +27,14 @@ argument_groups:
         name: --output_sc
         description: Processed single-cell dataset
         direction: output
-        required: true     
+        required: true
+        default: '$id/output_sc.h5ad'
       - type: file
         name: --output_sp
         description: Processed spatial dataset
         direction: output
         required: true
+        default: '$id/output_sp.h5ad'
 
   - name: Dataset metadata
     description: Metadata about the dataset, used for documentation and provenance.

src/process_datasets/sc_features/script.R

@@ -45,4 +45,4 @@ cat("Output dataset:\n")
 print(input_sp)
 
 cat("Write output files\n")
-output_sp$write_h5ad(par$output_sp, compression = "gzip")
+input_sp$write_h5ad(par$output_sp, compression = "gzip")

src/workflows/process_datasets/config.vsh.yaml

@@ -0,0 +1,18 @@
+__merge__: /src/api/comp_process_datasets.yaml
+name: "process_datasets"
+namespace: "workflows"
+resources:
+  - type: nextflow_script
+    path: main.nf
+    entrypoint: run_wf
+  - path: /common/nextflow_helpers/helper.nf
+dependencies:
+  - name: common/check_dataset_schema
+    repository: openproblems_v2
+  - name: common/extract_metadata
+    repository: openproblems_v2
+  - name: process_datasets/convert
+  - name: process_datasets/sc_features
+  - name: process_datasets/precompute_downstream
+runners:
+  - type: nextflow

src/workflows/process_datasets/main.nf

@@ -0,0 +1,54 @@
+include { findArgumentSchema } from "${meta.resources_dir}/helper.nf"
+
+workflow auto {
+  findStates(params, meta.config)
+    | meta.workflow.run(
+      auto: [publish: "state"]
+    )
+}
+
+workflow run_wf {
+  take:
+  input_ch
+
+  main:
+
+  output_ch = input_ch
+
+    | convert.run(
+      fromState: [
+        "input_sc",
+        "input_sp",
+        "dataset_id",
+        "dataset_name",
+        "dataset_url",
+        "dataset_reference",
+        "dataset_summary",
+        "dataset_description",
+        "dataset_organism"
+      ],
+      toState: [
+        "output_sc",
+        "output_sp"
+      ]
+    )
+
+    | sc_features.run(
+      fromState: ["input_sp": "output_sp"],
+      toState: ["output_sp"]
+    )
+
+    | precompute_downstream.run(
+      fromState: ["input_sp": "output_sp"],
+      toState: ["output_sp"]
+    )
+
+    | setState ([
+      "output_sc",
+      "output_sp",
+      "_meta"
+    ])
+
+  emit:
+  output_ch
+}