Alternative split api

README.md

@@ -51,7 +51,8 @@ flowchart TB
   comp_method[/"<a href='https://github.com/openproblems-bio/task_cyto_batch_integration#component-type-method'>Method</a>"/]
   comp_control_method[/"<a href='https://github.com/openproblems-bio/task_cyto_batch_integration#component-type-control-method'>Control Method</a>"/]
   comp_metric[/"<a href='https://github.com/openproblems-bio/task_cyto_batch_integration#component-type-metric'>Metric</a>"/]
-  file_integrated("<a href='https://github.com/openproblems-bio/task_cyto_batch_integration#file-format-integrated'>Integrated</a>")
+  file_integrated_split1("<a href='https://github.com/openproblems-bio/task_cyto_batch_integration#file-format-integrated'>Integrated</a>")
+  file_integrated_split2("<a href='https://github.com/openproblems-bio/task_cyto_batch_integration#file-format-integrated'>Integrated</a>")
   file_score("<a href='https://github.com/openproblems-bio/task_cyto_batch_integration#file-format-score'>Score</a>")
   file_common_dataset---comp_data_processor
   comp_data_processor-->file_censored
@@ -60,12 +61,12 @@ flowchart TB
   file_censored---comp_method
   file_unintegrated---comp_control_method
   file_unintegrated---comp_metric
-  comp_method-->file_integrated
-  comp_control_method-->file_integrated
-  comp_control_method-->file_integrated
+  comp_method-->file_integrated_split1
+  comp_control_method-->file_integrated_split1
+  comp_control_method-->file_integrated_split2
   comp_metric-->file_score
-  file_integrated---comp_metric
-  file_integrated---comp_metric
+  file_integrated_split1---comp_metric
+  file_integrated_split2---comp_metric
 ```
 
 ## File format: Common Dataset
@@ -131,8 +132,8 @@ Arguments:
 | Name | Type | Description |
 |:---|:---|:---|
 | `--input` | `file` | A subset of the common dataset. |
-| `--output_censored_left` | `file` | (*Output*) An unintegrated dataset with certain columns (cells metadata), such as the donor information, hidden. These columns are intentionally hidden to prevent bias. |
-| `--output_censored_right` | `file` | (*Output*) An unintegrated dataset with certain columns (cells metadata), such as the donor information, hidden. These columns are intentionally hidden to prevent bias. |
+| `--output_censored_split1` | `file` | (*Output*) An unintegrated dataset with certain columns (cells metadata), such as the donor information, hidden. These columns are intentionally hidden to prevent bias. |
+| `--output_censored_split2` | `file` | (*Output*) An unintegrated dataset with certain columns (cells metadata), such as the donor information, hidden. These columns are intentionally hidden to prevent bias. |
 | `--output_unintegrated` | `file` | (*Output*) The complete unintegrated dataset. |
 
 </div>
@@ -274,8 +275,8 @@ Arguments:
 | Name | Type | Description |
 |:---|:---|:---|
 | `--input_unintegrated` | `file` | The complete unintegrated dataset. |
-| `--output_integrated_left` | `file` | (*Output*) Integrated dataset which batch effect was corrected by an algorithm. |
-| `--output_integrated_right` | `file` | (*Output*) Integrated dataset which batch effect was corrected by an algorithm. |
+| `--output_integrated_split1` | `file` | (*Output*) Integrated dataset which batch effect was corrected by an algorithm. |
+| `--output_integrated_split2` | `file` | (*Output*) Integrated dataset which batch effect was corrected by an algorithm. |
 
 </div>
 
@@ -290,8 +291,8 @@ Arguments:
 | Name | Type | Description |
 |:---|:---|:---|
 | `--input_unintegrated` | `file` | The complete unintegrated dataset. |
-| `--input_integrated_left` | `file` | Integrated dataset which batch effect was corrected by an algorithm. |
-| `--input_integrated_right` | `file` | Integrated dataset which batch effect was corrected by an algorithm. |
+| `--input_integrated_split1` | `file` | Integrated dataset which batch effect was corrected by an algorithm. |
+| `--input_integrated_split2` | `file` | Integrated dataset which batch effect was corrected by an algorithm. |
 | `--output` | `file` | (*Output*) File indicating the score of a metric. |
 
 </div>
@@ -301,7 +302,37 @@ Arguments:
 Integrated dataset which batch effect was corrected by an algorithm
 
 Example file:
-`resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/integrated.h5ad`
+`resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/integrated_split1.h5ad`
+
+Format:
+
+<div class="small">
+
+    AnnData object
+     layers: 'integrated'
+     uns: 'dataset_id', 'method_id', 'parameters'
+
+</div>
+
+Data structure:
+
+<div class="small">
+
+| Slot | Type | Description |
+|:---|:---|:---|
+| `layers["integrated"]` | `double` | The integrated data as returned by a batch correction method. |
+| `uns["dataset_id"]` | `string` | A unique identifier for the dataset. |
+| `uns["method_id"]` | `string` | A unique identifier for the method. |
+| `uns["parameters"]` | `object` | (*Optional*) The parameters used for the integration. |
+
+</div>
+
+## File format: Integrated
+
+Integrated dataset which batch effect was corrected by an algorithm
+
+Example file:
+`resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/integrated_split2.h5ad`
 
 Format:
 

scripts/create_resources/process_raw_datasets.sh

@@ -4,7 +4,7 @@ cat > /tmp/params.yaml << 'HERE'
 input_states: s3://openproblems-data/resources/task_cyto_batch_integration/datasets_raw/**/state.yaml
 rename_keys: 'input:output_dataset'
 output_state: '$id/state.yaml'
-settings: '{"output_unintegrated": "$id/unintegrated.h5ad", "output_censored_left": "$id/censored_left.h5ad", "output_censored_right": "$id/censored_right.h5ad"}'
+settings: '{"output_unintegrated": "$id/unintegrated.h5ad", "output_censored_split1": "$id/censored_split1.h5ad", "output_censored_split2": "$id/censored_split2.h5ad"}'
 publish_dir: s3://openproblems-data/resources/task_cyto_batch_integration/datasets/
 HERE
 

scripts/create_test_resources/mouse_spleen_flow_cytometry_subset.sh

@@ -52,34 +52,34 @@ HERE
 viash run src/data_processors/process_dataset/config.vsh.yaml -- \
   --input $DATASET_DIR/common_dataset.h5ad \
   --output_unintegrated $DATASET_DIR/unintegrated.h5ad \
-  --output_censored_left $DATASET_DIR/censored_left.h5ad \
-  --output_censored_right $DATASET_DIR/censored_right.h5ad
+  --output_censored_split1 $DATASET_DIR/censored_split1.h5ad \
+  --output_censored_split2 $DATASET_DIR/censored_split2.h5ad
 
 # run one method
 viash run src/methods/harmonypy/config.vsh.yaml -- \
-  --input $DATASET_DIR/censored_left.h5ad \
-  --output $DATASET_DIR/integrated_left.h5ad
+  --input $DATASET_DIR/censored_split1.h5ad \
+  --output $DATASET_DIR/integrated_split1.h5ad
 
 # run one method
 viash run src/methods/harmonypy/config.vsh.yaml -- \
-  --input $DATASET_DIR/censored_right.h5ad \
-  --output $DATASET_DIR/integrated_right.h5ad
+  --input $DATASET_DIR/censored_split2.h5ad \
+  --output $DATASET_DIR/integrated_split2.h5ad
 
 # run one metric
 viash run src/metrics/emd/config.vsh.yaml -- \
     --input_unintegrated $DATASET_DIR/unintegrated.h5ad \
-    --input_integrated_left $DATASET_DIR/integrated_left.h5ad \
-    --input_integrated_right $DATASET_DIR/integrated_right.h5ad \
+    --input_integrated_split1 $DATASET_DIR/integrated_split1.h5ad \
+    --input_integrated_split2 $DATASET_DIR/integrated_split2.h5ad \
     --output $DATASET_DIR/score.h5ad
 
 # write manual state.yaml
 cat > $DATASET_DIR/state.yaml << HERE
 id: $DATASET_ID
 unintegrated: !file unintegrated.h5ad
-censored_left: !file censored_left.h5ad
-censored_right: !file censored_right.h5ad
-integrated_left: !file integrated_left.h5ad
-integrated_right: !file integrated_right.h5ad
+censored_split1: !file censored_split1.h5ad
+censored_split2: !file censored_split2.h5ad
+integrated_split1: !file integrated_split1.h5ad
+integrated_split2: !file integrated_split2.h5ad
 score: !file score.h5ad
 HERE
 

scripts/run_benchmark/run_full_local.sh

@@ -18,7 +18,7 @@ publish_dir="resources/results/${RUN_ID}"
 # write the parameters to file
 cat > /tmp/params.yaml << HERE
 input_states: resources/datasets/**/state.yaml
-rename_keys: 'input_censored_left:output_censored_left;input_censored_right:output_censored_right;input_unintegrated:output_unintegrated'
+rename_keys: 'input_censored_split1:output_censored_split1;input_censored_split2:output_censored_split2;input_unintegrated:output_unintegrated'
 output_state: "state.yaml"
 publish_dir: "$publish_dir"
 HERE

scripts/run_benchmark/run_full_seqeracloud.sh

@@ -15,7 +15,7 @@ publish_dir="s3://openproblems-data/resources/task_cyto_batch_integration/result
 # write the parameters to file
 cat > /tmp/params.yaml << HERE
 input_states: s3://openproblems-data/resources/task_cyto_batch_integration/datasets/**/state.yaml
-rename_keys: 'input_censored_left:output_censored_left;input_censored_right:output_censored_right;input_unintegrated:output_unintegrated'
+rename_keys: 'input_censored_split1:output_censored_split1;input_censored_split2:output_censored_split2;input_unintegrated:output_unintegrated'
 output_state: "state.yaml"
 publish_dir: "$publish_dir"
 HERE

scripts/run_benchmark/run_test_local.sh

@@ -22,7 +22,7 @@ nextflow run . \
   -c common/nextflow_helpers/labels_ci.config \
   --id mouse_spleen_flow_cytometry_subset \
   --input_unintegrated resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/unintegrated.h5ad \
-  --input_censored_left resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/censored_left.h5ad \
-  --input_censored_right resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/censored_right.h5ad \
+  --input_censored_split1 resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/censored_split1.h5ad \
+  --input_censored_split2 resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/censored_split2.h5ad \
   --output_state state.yaml \
   --publish_dir "$publish_dir"

src/api/comp_control_method.yaml

@@ -16,12 +16,12 @@ arguments:
     __merge__: file_unintegrated.yaml
     required: true
     direction: input
-  - name: --output_integrated_left
-    __merge__: file_integrated.yaml
+  - name: --output_integrated_split1
+    __merge__: file_integrated_split1.yaml
     required: true
     direction: output
-  - name: --output_integrated_right
-    __merge__: file_integrated.yaml
+  - name: --output_integrated_split2
+    __merge__: file_integrated_split2.yaml
     required: true
     direction: output
 test_resources:

src/api/comp_data_processor.yaml

@@ -11,11 +11,11 @@ arguments:
     __merge__: file_common_dataset.yaml
     direction: input
     required: true
-  - name: "--output_censored_left"
+  - name: "--output_censored_split1"
     __merge__: file_censored.yaml
     direction: output
     required: true
-  - name: "--output_censored_right"
+  - name: "--output_censored_split2"
     __merge__: file_censored.yaml
     direction: output
     required: true

src/api/comp_method.yaml

@@ -12,7 +12,7 @@ arguments:
     required: true
     direction: input
   - name: --output
-    __merge__: file_integrated.yaml
+    __merge__: file_integrated_split1.yaml
     required: true
     direction: output
 test_resources:

src/api/comp_metric.yaml

@@ -11,12 +11,12 @@ arguments:
     __merge__: file_unintegrated.yaml
     direction: input
     required: true
-  - name: "--input_integrated_left"
-    __merge__: file_integrated.yaml
+  - name: "--input_integrated_split1"
+    __merge__: file_integrated_split1.yaml
     direction: input
     required: true
-  - name: "--input_integrated_right"
-    __merge__: file_integrated.yaml
+  - name: "--input_integrated_split2"
+    __merge__: file_integrated_split2.yaml
     direction: input
     required: true
   - name: "--output"

src/api/file_integrated.yaml → src/api/file_integrated_split1.yaml

@@ -1,5 +1,5 @@
 type: file
-example: "resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/integrated_left.h5ad"
+example: "resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/integrated_split1.h5ad"
 label: Integrated
 summary: "Integrated dataset which batch effect was corrected by an algorithm"
 info:

src/api/file_integrated_split2.yaml

@@ -0,0 +1,25 @@
+type: file
+example: "resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/integrated_split2.h5ad"
+label: Integrated
+summary: "Integrated dataset which batch effect was corrected by an algorithm"
+info:
+  format:
+    type: h5ad
+    layers:
+      - type: double
+        name: integrated
+        description: The integrated data as returned by a batch correction method
+        required: true
+    uns:
+      - type: string
+        name: dataset_id
+        description: "A unique identifier for the dataset"
+        required: true
+      - type: string
+        name: method_id
+        description: "A unique identifier for the method"
+        required: true
+      - type: object
+        name: parameters
+        description: "The parameters used for the integration"
+        required: false

src/control_methods/no_integration/script.py

@@ -4,8 +4,8 @@
 # The following code has been auto-generated by Viash.
 par = {
   'input_unintegrated': r'resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/unintegrated.h5ad',
-  'output_integrated_left': r'resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/integrated.h5ad',
-  'output_integrated_right': r'resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/integrated.h5ad'
+  'output_integrated_split1': r'resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/integrated.h5ad',
+  'output_integrated_split2': r'resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/integrated.h5ad'
 }
 meta = {
   'name': r'no_integration',
@@ -18,35 +18,35 @@
 
 print("Extracting and splitting unintegrated data", flush=True)
 #split 1
-adata_left = adata[(adata.obs.is_control>0) | (adata.obs.split==1)]
-integrated_left = adata_left.layers["preprocessed"]
+adata_split1 = adata[(adata.obs.is_control>0) | (adata.obs.split==1)]
+integrated_split1 = adata_split1.layers["preprocessed"]
 #split 2
-adata_right = adata[(adata.obs.is_control>0) | (adata.obs.split==2)]
-integrated_right = adata_right.layers["preprocessed"]
+adata_split2 = adata[(adata.obs.is_control>0) | (adata.obs.split==2)]
+integrated_split2 = adata_split2.layers["preprocessed"]
 
 print("Write output AnnData to files", flush=True)
 #split 1
-output_left = ad.AnnData(
-    obs=adata_left.obs[[]],
-    var=adata_left.var[[]],
-    layers={"integrated": integrated_left},
+output_split1 = ad.AnnData(
+    obs=adata_split1.obs[[]],
+    var=adata_split1.var[[]],
+    layers={"integrated": integrated_split1},
     uns={
         "dataset_id": adata.uns["dataset_id"],
         "method_id": meta["name"],
         "parameters": {},
     },
 )
 #split 2
-output_right = ad.AnnData(
-    obs=adata_right.obs[[]],
-    var=adata_right.var[[]],
-    layers={"integrated": integrated_right},
+output_split2 = ad.AnnData(
+    obs=adata_split2.obs[[]],
+    var=adata_split2.var[[]],
+    layers={"integrated": integrated_split2},
     uns={
         "dataset_id": adata.uns["dataset_id"],
         "method_id": meta["name"],
         "parameters": {},
     },
 )
 
-output_left.write_h5ad(par["output_integrated_left"], compression="gzip")
-output_right.write_h5ad(par["output_integrated_right"], compression="gzip")
+output_split1.write_h5ad(par["output_integrated_split1"], compression="gzip")
+output_split2.write_h5ad(par["output_integrated_split2"], compression="gzip")

src/control_methods/perfect_integration/script.py

@@ -4,8 +4,8 @@
 # The following code has been auto-generated by Viash.
 par = {
     "input_unintegrated": "resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/unintegrated.h5ad",
-    "output_integrated_left": "resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/integrated_left.h5ad",
-    "output_integrated_right": "resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/integrated_right.h5ad",
+    "output_integrated_split1": "resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/integrated_split1.h5ad",
+    "output_integrated_split2": "resources_test/task_cyto_batch_integration/mouse_spleen_flow_cytometry_subset/integrated_split2.h5ad",
 }
 meta = {"name": "perfect_integration"}
 
@@ -17,34 +17,34 @@
 print("Extracting and splitting unintegrated data", flush=True)
 
 # split 1
-adata_left = adata[(adata.obs.is_control > 0) | (adata.obs.batch == 1)]
-integrated_left = adata_left.layers["preprocessed"]
+adata_split1 = adata[(adata.obs.is_control > 0) | (adata.obs.batch == 1)]
+integrated_split1 = adata_split1.layers["preprocessed"]
 
 # split 2 == split 1 in this case
 
 print("Write output AnnData to file", flush=True)
 # split 1
-output_left = ad.AnnData(
-    obs=adata_left.obs[[]],
-    var=adata_left.var[[]],
-    layers={"integrated": integrated_left},
+output_split1 = ad.AnnData(
+    obs=adata_split1.obs[[]],
+    var=adata_split1.var[[]],
+    layers={"integrated": integrated_split1},
     uns={
         "dataset_id": adata.uns["dataset_id"],
         "method_id": meta["name"],
         "parameters": {},
     },
 )
 # split 2
-output_right = ad.AnnData(
-    obs=adata_left.obs[[]],
-    var=adata_left.var[[]],
-    layers={"integrated": integrated_left},
+output_split2 = ad.AnnData(
+    obs=adata_split1.obs[[]],
+    var=adata_split1.var[[]],
+    layers={"integrated": integrated_split1},
     uns={
         "dataset_id": adata.uns["dataset_id"],
         "method_id": meta["name"],
         "parameters": {},
     },
 )
 
-output_left.write_h5ad(par["output_integrated_left"], compression="gzip")
-output_right.write_h5ad(par["output_integrated_right"], compression="gzip")
+output_split1.write_h5ad(par["output_integrated_split1"], compression="gzip")
+output_split2.write_h5ad(par["output_integrated_split2"], compression="gzip")