Merge pull request #34 from openproblems-bio/benchmark_run

Benchmark run

Author: LouisK92

scripts/create_resources/README.md

@@ -0,0 +1,17 @@
+
+
+
+The spatial datasets can be downloaded and processed with scripts in `task_ist_preprocessing/scripts/create_resources/spatial` 
+
+The scRNAseq reference datasets can be downloaded and processed with scripts in `task_ist_preprocessing/scripts/create_resources/sc` 
+
+Note that scripts in those two dirs create the datasets at:  
+```
+s3://openproblems-data/resources/datasets/*/
+```
+
+Finally the script `task_ist_preprocessing/scripts/create_resources/combine/process_datasets.sh` creates the combined, processed datasets (pairs of spatial and scRNAseq). This processing/combining step subsets the genes to the common genes and saves metadata of the combined object in the output files. The combined datasets are saved at 
+```
+s3://openproblems-data/resources/task_ist_preprocessing/datasets/
+```
+(The final benchmark runs on all datasets in this location)

…pts/create_resources/process_datasets.sh …te_resources/combine/process_datasets.shscripts/create_resources/process_datasets.sh renamed to scripts/create_resources/combine/process_datasets.sh scripts/create_resources/process_datasets.sh renamed to scripts/create_resources/combine/process_datasets.sh

@@ -39,6 +39,6 @@ tw launch https://github.com/openproblems-bio/task_ist_preprocessing.git \
   --config common/nextflow_helpers/labels_tw.config \
   --labels datasets,wu_human_breast_cancer_sc
 
-aws s3 sync \
-  s3://openproblems-data/resources/datasets/wu_human_breast_cancer_sc/2021Wu_human_breast_cancer_sc \
-  resources/datasets/wu_human_breast_cancer_sc/2021Wu_human_breast_cancer_sc
+#aws s3 sync \
+#  s3://openproblems-data/resources/datasets/wu_human_breast_cancer_sc/2021Wu_human_breast_cancer_sc \
+#  resources/datasets/wu_human_breast_cancer_sc/2021Wu_human_breast_cancer_sc

…/process_allen_brain_cell_atlas_brain.sh …/process_allen_brain_cell_atlas_brain.shscripts/create_resources/process_allen_brain_cell_atlas_brain.sh renamed to scripts/create_resources/sc/process_allen_brain_cell_atlas_brain.sh scripts/create_resources/process_allen_brain_cell_atlas_brain.sh renamed to scripts/create_resources/sc/process_allen_brain_cell_atlas_brain.sh

@@ -0,0 +1,11 @@
+
+
+Here we generate a small test dataset, used for `viash test`. Note that the file structure here is a bit simplified compared to `scripts/create_resources` as we only have one dataset.
+
+Download and process the single cell data:
+`bash 2023_yao_mouse_brain_scrnaseq_10xv2.sh`
+
+Download and process the spatial data:
+`bash 2023_10x_mouse_brain_xenium_rep1.sh`
+
+Combine the two datasets and run the ist preprocessing pipeline once with generic methods to create example outputs after each step.

…ces/process_wu_human_breast_cancer_sc.sh …/sc/process_wu_human_breast_cancer_sc.shscripts/create_resources/process_wu_human_breast_cancer_sc.sh renamed to scripts/create_resources/sc/process_wu_human_breast_cancer_sc.sh scripts/create_resources/process_wu_human_breast_cancer_sc.sh renamed to scripts/create_resources/sc/process_wu_human_breast_cancer_sc.sh

@@ -27,7 +27,7 @@ tw launch https://github.com/openproblems-bio/task_ist_preprocessing.git \
   --workspace 53907369739130 \
   --params-file /tmp/params.yaml \
   --config common/nextflow_helpers/labels_tw.config \
-  --labels task_template,test
+  --labels task_ist_preprocessing,test
 
 aws s3 sync \
   s3://openproblems-nextflow/temp/results \

…s/create_resources/process_10x_xenium.sh …_resources/spatial/process_10x_xenium.shscripts/create_resources/process_10x_xenium.sh renamed to scripts/create_resources/spatial/process_10x_xenium.sh scripts/create_resources/process_10x_xenium.sh renamed to scripts/create_resources/spatial/process_10x_xenium.sh

@@ -397,8 +397,8 @@ workflow run_wf {
           // dataset_sp_id: ..., // todo: extract this from the dataset
           method_ids: method_ids,
           steps: state.steps,
-          metric_ids: state.score_uns.metrics_ids,
-          metrics_values: state.score_uns.metrics_values
+          metric_ids: state.score_uns.metric_ids,
+          metrics_values: state.score_uns.metric_values
         ]
       }
       def score_uns_yaml_blob = toYamlBlob(score_uns)