restructuring data storage and local workflows

README.md

@@ -71,7 +71,7 @@ flowchart TB
 
 Chromatin accessibility data
 
-Example file: `resources_test/inference_datasets/op_atac.h5ad`
+Example file: `resources_test/grn_benchmark/inference_datasets//op_atac.h5ad`
 
 Format:
 
@@ -245,7 +245,7 @@ Data structure:
 
 Perturbation dataset for benchmarking.
 
-Example file: `resources_test/evaluation_datasets/op_perturbation.h5ad`
+Example file: `resources_test/grn_benchmark/evaluation_datasets//op_perturbation.h5ad`
 
 Format:
 
@@ -275,7 +275,7 @@ Data structure:
 
 RNA expression data.
 
-Example file: `resources_test/inference_datasets/op_rna.h5ad`
+Example file: `resources_test/grn_benchmark/inference_datasets//op_rna.h5ad`
 
 Format:
 

_viash.yaml

@@ -30,10 +30,10 @@ info:
   test_resources:
     - type: s3
       path: s3://openproblems-data/resources_test/grn/inference_datasets/
-      dest: resources_test/inference_datasets/
+      dest: resources_test/grn_benchmark/inference_datasets//
     - type: s3
       path: s3://openproblems-data/resources_test/grn/evaluation_datasets/
-      dest: resources_test/evaluation_datasets/
+      dest: resources_test/grn_benchmark/evaluation_datasets//
     - type: s3
       path: s3://openproblems-data/resources_test/grn/prior/
       dest: resources_test/prior/
@@ -67,7 +67,7 @@ info:
 
     ```bash
     viash run src/control_methods/pearson_corr/config.vsh.yaml -- \
-              --rna resources/inference_datasets/norman_rna.h5ad --prediction output/net.h5ad
+              --rna resources/grn_benchmark/inference_datasets/norman_rna.h5ad --prediction output/net.h5ad
     ```
 
     ## Evaluate a GRN

scripts/add_a_method.sh

@@ -27,7 +27,7 @@ viash run src/methods/$method_id/config.vsh.yaml -- \
 
 # run the inference using the method for op dataset using only RNA data. Add more aurguments if needed.
 viash run src/methods/$method_id/config.vsh.yaml -- \
-  --rna "resources/inference_datasets/op_rna.h5ad" \
+  --rna "resources/grn_benchmark/inference_datasets/op_rna.h5ad" \
   --prediction "output/prediction.h5ad"
 
 # run evaluation metrics

scripts/single_grn_evaluation.sh

@@ -7,9 +7,9 @@ viash run src/metrics/all_metrics/config.novsh.yaml -- \
     --prediction ${prediction} \
     --dataset_id ${dataset_id} \
     --score output/score.h5ad \
-    --tf_all resources/prior/tf_all.csv \
-    --regulators_consensus resources/prior/regulators_consensus_${dataset_id}.json \
-    --ws_consensus resources/prior/ws_consensus_${dataset_id}.csv \
-    --ws_distance_background resources/prior/ws_distance_background_${dataset_id}.csv \
-    --evaluation_data_sc resources/evaluation_datasets/${dataset_id}_sc_counts.h5ad \
-    --evaluation_data resources/evaluation_datasets/${dataset_id}_perturbation.h5ad
+    --tf_all resources/grn_benchmark/prior/tf_all.csv \
+    --regulators_consensus resources/grn_benchmark/prior/regulators_consensus_${dataset_id}.json \
+    --ws_consensus resources/grn_benchmark/prior/ws_consensus_${dataset_id}.csv \
+    --ws_distance_background resources/grn_benchmark/prior/ws_distance_background_${dataset_id}.csv \
+    --evaluation_data_sc resources/grn_benchmark/evaluation_datasets//${dataset_id}_sc_counts.h5ad \
+    --evaluation_data resources/grn_benchmark/evaluation_datasets//${dataset_id}_perturbation.h5ad

src/api/file_atac_h5ad.yaml

@@ -1,5 +1,5 @@
 type: file
-example: resources_test/inference_datasets/op_atac.h5ad
+example: resources_test/grn_benchmark/inference_datasets//op_atac.h5ad
 label:  chromatin accessibility data
 summary: "Chromatin accessibility data"
 info:

src/api/file_evaluation_h5ad.yaml

@@ -1,5 +1,5 @@
 type: file
-example: resources_test/evaluation_datasets/op_perturbation.h5ad
+example: resources_test/grn_benchmark/evaluation_datasets/op_bulk.h5ad
 label: perturbation data
 summary: "Perturbation dataset for benchmarking."
 

src/api/file_prediction_h5ad.yaml

@@ -15,7 +15,7 @@ info:
         name: method_id
         description: "A unique identifier for the inference method"
         required: true
-      - type: DataFrame
+      - type: object
         name: prediction
         description: "Inferred GRNs in the format of source, target, weight"
         required: true

src/api/file_rna_h5ad.yaml

@@ -1,5 +1,5 @@
 type: file
-example: resources_test/inference_datasets/op_rna.h5ad
+example: resources_test/grn_benchmark/inference_datasets/op_rna.h5ad
 label: gene expression data
 summary: "RNA expression data."
 info:

src/control_methods/negative_control/script.py

@@ -8,7 +8,7 @@
 par = {
   "rna": "resources/grn-benchmark/rna.h5ad",
   "prediction": "resources/grn_models/default/negative_control.csv",
-  "tf_all": "resources/prior/tf_all.csv",
+  "tf_all": "resources/grn_benchmark/prior/tf_all.csv",
   "max_n_links": 50000
 }
 ## VIASH END

src/control_methods/pearson_corr/script.py

@@ -10,8 +10,8 @@
 
 ## VIASH START
 par = {
-    'rna': 'resources/evaluation_datasets/op_rna.h5ad',
-    'tf_all': 'resources/prior/tf_all.csv',
+    'rna': 'resources/grn_benchmark/evaluation_datasets//op_rna.h5ad',
+    'tf_all': 'resources/grn_benchmark/prior/tf_all.csv',
     'cell_type_specific': False,
     'max_n_links': 50000,
     'prediction': 'output/pearson_net.h5ad',

src/control_methods/pearson_corr/test.sh

@@ -1,4 +1,4 @@
 viash run src/control_methods/pearson/config.vsh.yaml -- \
   --prediction output/baseline_corr.csv \
   --multiomics_rna resources/grn-benchmark/multiomics_rna.h5ad \
-  --tf_all resources/prior/tf_all.csv 
+  --tf_all resources/grn_benchmark/prior/tf_all.csv 

src/control_methods/positive_control/script.py

@@ -12,7 +12,7 @@
 ## VIASH START
 par = {
     'rna': 'resources/grn-benchmark/rna.h5ad',
-    'tf_all': 'resources/prior/tf_all.csv',
+    'tf_all': 'resources/grn_benchmark/prior/tf_all.csv',
     'max_n_links': 50000,
     'prediction': 'resources/grn_models/positive_control.csv',
     "seed": 32,

src/control_methods/positive_control/test.sh

@@ -1,5 +1,5 @@
 viash run src/control_methods/baseline_corr/config.vsh.yaml -- \
   --prediction output/baseline_corr.csv \
   --multiomics_rna resources/grn-benchmark/multiomics_rna.h5ad \
-  --tf_all resources/prior/tf_all.csv \
+  --tf_all resources/grn_benchmark/prior/tf_all.csv \
   --causal true

src/helper.py

@@ -40,12 +40,12 @@ def analyse_meta_cells(task_grn_inference_dir):
 
 
     par = {
-        'rna': f'{task_grn_inference_dir}/resources/inference_datasets/{dataset}_rna.h5ad',
-        "evaluation_data": f"{task_grn_inference_dir}/resources/evaluation_datasets/{dataset}_perturbation.h5ad",
+        'rna': f'{task_grn_inference_dir}/resources/grn_benchmark/inference_datasets/{dataset}_rna.h5ad',
+        "evaluation_data": f"{task_grn_inference_dir}/resources/grn_benchmark/evaluation_datasets//{dataset}_perturbation.h5ad",
 
         'layer': 'X_norm',
-        'consensus':  f'{task_grn_inference_dir}/resources/prior/{dataset}_consensus-num-regulators.json',
-        'tf_all': f'{task_grn_inference_dir}/resources/prior/tf_all.csv',
+        'consensus':  f'{task_grn_inference_dir}/resources/grn_benchmark/prior/{dataset}_consensus-num-regulators.json',
+        'tf_all': f'{task_grn_inference_dir}/resources/grn_benchmark/prior/tf_all.csv',
         'apply_tf': True,
         'apply_skeleton': False,
         'verbose': 2,
@@ -123,12 +123,12 @@ def analyse_imputation(task_grn_inference_dir):
 
 
     par = {
-        'rna': f'{task_grn_inference_dir}/resources/inference_datasets/{dataset}_rna.h5ad',
-        "evaluation_data": f"{task_grn_inference_dir}/resources/evaluation_datasets/{dataset}_perturbation.h5ad",
+        'rna': f'{task_grn_inference_dir}/resources/grn_benchmark/inference_datasets/{dataset}_rna.h5ad',
+        "evaluation_data": f"{task_grn_inference_dir}/resources/grn_benchmark/evaluation_datasets//{dataset}_perturbation.h5ad",
 
         'layer': 'X_norm',
-        'consensus':  f'{task_grn_inference_dir}/resources/prior/{dataset}_consensus-num-regulators.json',
-        'tf_all': f'{task_grn_inference_dir}/resources/prior/tf_all.csv',
+        'consensus':  f'{task_grn_inference_dir}/resources/grn_benchmark/prior/{dataset}_consensus-num-regulators.json',
+        'tf_all': f'{task_grn_inference_dir}/resources/grn_benchmark/prior/tf_all.csv',
         'apply_tf': True,
         'apply_skeleton': False,
         'verbose': 2,
@@ -204,12 +204,12 @@ def analyse_imputation(task_grn_inference_dir):
 def analyse_corr_vs_tfmasked_corr(task_grn_inference_dir):
     for i_run, dataset in enumerate(['op', 'replogle2', 'nakatake', 'norman', 'adamson']):
         par = {
-            'rna': f'{task_grn_inference_dir}/resources/inference_datasets/{dataset}_rna.h5ad',
-            "evaluation_data": f"{task_grn_inference_dir}/resources/evaluation_datasets/{dataset}_perturbation.h5ad",
+            'rna': f'{task_grn_inference_dir}/resources/grn_benchmark/inference_datasets/{dataset}_rna.h5ad',
+            "evaluation_data": f"{task_grn_inference_dir}/resources/grn_benchmark/evaluation_datasets//{dataset}_perturbation.h5ad",
 
             'layer': 'X_norm',
-            'consensus':  f'{task_grn_inference_dir}/resources/prior/{dataset}_consensus-num-regulators.json',
-            'tf_all': f'{task_grn_inference_dir}/resources/prior/tf_all.csv',
+            'consensus':  f'{task_grn_inference_dir}/resources/grn_benchmark/prior/{dataset}_consensus-num-regulators.json',
+            'tf_all': f'{task_grn_inference_dir}/resources/grn_benchmark/prior/tf_all.csv',
             'apply_skeleton': False,
             'verbose': 2,
             'max_n_links': 50_000,
@@ -342,7 +342,7 @@ def create_skeleton():
     skeleton = skeleton.drop_duplicates().reset_index(drop=True)
     skeleton.nunique()
     all_links = skeleton['source'].astype(str) + '_' + skeleton['target'].astype(str)
-    np.savetxt('resources/prior/skeleton.csv', all_links.values, fmt='%s')
+    np.savetxt('resources/grn_benchmark/prior/skeleton.csv', all_links.values, fmt='%s')
 
 def merge_tf_motifs():
     cols = ['chrom', 'chromStart',  'chromEnd', 'name']

src/methods/multi_omics/celloracle/script.py

@@ -6,8 +6,8 @@
 
 ## VIASH START
 par = {
-  "rna": "resources/inference_datasets/op_rna.h5ad",
-  "atac": "resources/inference_datasets/op_atac.h5ad",
+  "rna": "resources/grn_benchmark/inference_datasets/op_rna.h5ad",
+  "atac": "resources/grn_benchmark/inference_datasets/op_atac.h5ad",
   "base_grn": 'output/celloracle/base_grn.csv',
   "temp_dir": 'output/celloracle/',
   "num_workers": 10,

src/methods/multi_omics/figr/script.R

@@ -11,7 +11,7 @@ par <- list(
   multiomics_rna_r = "resources/grn-benchmark/multiomics_rna.rds",
   multiomics_atac_r = "resources/grn-benchmark/multiomics_atac.rds",
   temp_dir =  "output/figr/",
-  cell_topic = "resources/prior/cell_topic.csv",
+  cell_topic = "resources/grn_benchmark/prior/cell_topic.csv",
   num_workers = 20,
   n_topics = 48,
   peak_gene = "output/figr/peak_gene.csv",

src/methods/multi_omics/scenicplus/test.sh

@@ -1,2 +1,2 @@
-viash run src/methods/multi_omics/scenicplus/config.vsh.yaml -- --multiomics_atac ./resources_test/grn-benchmark/multiomics_atac.h5ad --multiomics_rna ./resources_test/grn-benchmark/multiomics_rna.h5ad --tf_all ./resources/prior/tf_all.csv --prediction ./output/scenicplus/prediction.csv --cell_topic ./output/scenicplus/cell_topic.csv --scplus_mdata ./output/scenicplus/scplus_mdata.h5mu --temp_dir ./output/scenicplus/
+viash run src/methods/multi_omics/scenicplus/config.vsh.yaml -- --multiomics_atac ./resources_test/grn-benchmark/multiomics_atac.h5ad --multiomics_rna ./resources_test/grn-benchmark/multiomics_rna.h5ad --tf_all ./resources/grn_benchmark/prior/tf_all.csv --prediction ./output/scenicplus/prediction.csv --cell_topic ./output/scenicplus/cell_topic.csv --scplus_mdata ./output/scenicplus/scplus_mdata.h5mu --temp_dir ./output/scenicplus/
 #viash run src/methods/multi_omics/scenicplus/config.vsh.yaml -p docker -- ---setup build

src/methods/single_omics/ennet/script.R

@@ -5,7 +5,7 @@ library(dplyr)
 ## VIASH START
 par <- list(
     "multiomics_rna" = 'resources/resources_test/grn-benchmark/multiomics_rna.h5ad',
-    "tf_all" = 'resources/prior/tf_all.csv',
+    "tf_all" = 'resources/grn_benchmark/prior/tf_all.csv',
     "prediction" = 'output/ennet/prediction.csv',
     "temp_dir": 'output/ennet',
     "max_n_links": 50000,

src/methods/single_omics/ennet/test.sh

@@ -1 +1 @@
-viash run src/methods/single_omics/ennet/config.vsh.yaml -- --multiomics_rna resources_test/grn-benchmark/multiomics_rna.h5ad --tf_all resources/prior/tf_all.csv --prediction output/ennet/prediction.csv
+viash run src/methods/single_omics/ennet/config.vsh.yaml -- --multiomics_rna resources_test/grn-benchmark/multiomics_rna.h5ad --tf_all resources/grn_benchmark/prior/tf_all.csv --prediction output/ennet/prediction.csv

src/methods/single_omics/genie3/script.py

@@ -12,7 +12,7 @@
 ## VIASH START
 par = {
   'rna': 'resources/grn-benchmark/rna_d0_hvg.h5ad',
-  "tf_all": 'resources/prior/tf_all.csv',
+  "tf_all": 'resources/grn_benchmark/prior/tf_all.csv',
   'prediction': 'output/genie3_donor0_hvg.csv',
   'max_n_links': 50000,
   'normalize': False

src/methods/single_omics/genie3/test.sh

@@ -1 +1 @@
-viash run src/methods/single_omics/genie3/config.novsh.yaml -- --multiomics_rna resources_test/grn-benchmark/multiomics_rna.h5ad --tfs resources/prior/tf_all.csv --prediction output/genie3/prediction.csv
+viash run src/methods/single_omics/genie3/config.novsh.yaml -- --multiomics_rna resources_test/grn-benchmark/multiomics_rna.h5ad --tfs resources/grn_benchmark/prior/tf_all.csv --prediction output/genie3/prediction.csv

src/methods/single_omics/grnboost2/script.py

@@ -9,7 +9,7 @@
 ## VIASH START
 par = {
   'rna': 'resources/grn-benchmark/rna_d0_hvg.h5ad',
-  "tf_all": 'resources/prior/tf_all.csv',
+  "tf_all": 'resources/grn_benchmark/prior/tf_all.csv',
   'prediction': 'output/grnboost2_donor_0_hvg.csv',
   'max_n_links': 50000,
   'cell_type_specific': False,

src/methods/single_omics/grnboost2/test.sh

@@ -1,4 +1,4 @@
 viash run src/methods/single_omics/grnboost2/config.vsh.yaml -- --multiomics_rna resources_test/grn-benchmark/multiomics_rna.h5ad \
-    --tf_all resources/prior/tf_all.csv \
+    --tf_all resources/grn_benchmark/prior/tf_all.csv \
     --cell_type_specific false \
     --prediction output/grnboost2/prediction.csv

src/methods/single_omics/portia/script.py

@@ -12,7 +12,7 @@
 ## VIASH START
 par = {
   'rna': 'resources/grn-benchmark/rna.h5ad',
-  'tf_all': 'resources/prior/tf_all.csv',
+  'tf_all': 'resources/grn_benchmark/prior/tf_all.csv',
   'prediction': 'output/portia.csv',
   'max_n_links': 50000,
   'donor_specific': False,

src/methods/single_omics/portia/test.sh

@@ -1,4 +1,4 @@
 viash run src/methods/single_omics/portia/config.vsh.yaml -- --multiomics_rna resources_test/grn-benchmark/multiomics_rna.h5ad \
-    --tf_all resources/prior/tf_all.csv \
+    --tf_all resources/grn_benchmark/prior/tf_all.csv \
     --prediction output/portia/prediction.csv \
     --cell_type_specific false

src/methods/single_omics/ppcor/script.R

@@ -6,7 +6,7 @@ library(dplyr)
 par <- list(
   rna = 'resources/grn-benchmark/rna.h5ad',
   prediction = 'resources/grn_models/ppcor.csv',
-  tf_all = 'resources/prior/tf_all.csv',
+  tf_all = 'resources/grn_benchmark/prior/tf_all.csv',
   max_n_links = 50000
 )
 ## VIASH END

src/methods/single_omics/scenic/script.py

@@ -19,7 +19,7 @@
 ## VIASH START
 par = {
   'rna': 'resources/grn-benchmark/rna_0.h5ad',
-  "tf_all": 'resources/prior/tf_all.csv',
+  "tf_all": 'resources/grn_benchmark/prior/tf_all.csv',
   'prediction': 'output/scenic_0_hvgs.csv',
   'temp_dir': 'output/scenic',
   'num_workers': 20,

src/methods/single_omics/scenic/test.sh

@@ -1 +1 @@
-viash run src/methods/single_omics/scenic/config.vsh.yaml -- --multiomics_rna resources_test/grn-benchmark/multiomics_rna.h5ad --tf_all resources/prior/tf_all.csv --prediction output/scenic_prediction.csv --temp_dir output/scenic
+viash run src/methods/single_omics/scenic/config.vsh.yaml -- --multiomics_rna resources_test/grn-benchmark/multiomics_rna.h5ad --tf_all resources/grn_benchmark/prior/tf_all.csv --prediction output/scenic_prediction.csv --temp_dir output/scenic

src/methods/single_omics/scgpt/run.sh

@@ -1,4 +1,4 @@
 viash run src/methods/single_omics/scgpt/config.vsh.yaml -- \
-    --rna resources_test/inference_datasets/op_rna.h5ad \
-    --tf_all resources/prior/tf_all.csv \
+    --rna resources_test/grn_benchmark/inference_datasets//op_rna.h5ad \
+    --tf_all resources/grn_benchmark/prior/tf_all.csv \
     --prediction output/prediction.h5ad

src/methods/single_omics/scgpt/test.sh

@@ -1 +1 @@
-viash run src/methods/single_omics/portia/config.novsh.yaml -- --multiomics_rna resources/grn-benchmark/multiomics_rna.h5ad --tfs resources/prior/tf_all.csv --prediction output/portia/prediction.csv
+viash run src/methods/single_omics/portia/config.novsh.yaml -- --multiomics_rna resources/grn-benchmark/multiomics_rna.h5ad --tfs resources/grn_benchmark/prior/tf_all.csv --prediction output/portia/prediction.csv

src/methods/single_omics/scprint/run.sh

@@ -1,10 +1,31 @@
+#!/bin/bash
+#SBATCH --job-name=scprint
+#SBATCH --output=logs/%j.out
+#SBATCH --error=logs/%j.err
+#SBATCH --ntasks=1
+#SBATCH --cpus-per-task=10
+#SBATCH --time=10:00:00
+#SBATCH --mem=100GB
+#SBATCH --partition=cpu
+#SBATCH --mail-type=END,FAIL      
+#SBATCH [email protected]   
+
+
+
 # viash run src/methods/single_omics/scprint/config.vsh.yaml -- \
-#     --rna resources_test/inference_datasets/op_rna.h5ad \
-#     --tf_all resources/prior/tf_all.csv \
+#     --rna resources_test/grn_benchmark/inference_datasets//op_rna.h5ad \
+#     --tf_all resources/grn_benchmark/prior/tf_all.csv \
+#     --prediction output/prediction.h5ad
+
+
+# python src/methods/single_omics/scprint/script.py  \
+#     --rna resources/grn_benchmark/inference_datasets/op_rna.h5ad \
+#     --tf_all resources/grn_benchmark/prior/tf_all.csv \
 #     --prediction output/prediction.h5ad
 
+# Exit immediately if a command exits with a non-zero status
+set -e
+source ~/miniconda3/bin/activate scprint
 
-python src/methods/single_omics/scprint/script.py  \
-    --rna resources/inference_datasets/op_rna.h5ad \
-    --tf_all resources/prior/tf_all.csv \
-    --prediction output/prediction.h5ad
+python src/methods/script_all.py
+python src/metrics/script_all_experiment.py

src/methods/single_omics/scprint/script.py

@@ -28,8 +28,8 @@
 
 ## VIASH START
 par = {
-    'rna': 'resources/inference_datasets/op_rna.h5ad',
-    'tf_all': 'resources/prior/tf_all.csv',
+    'rna': 'resources/grn_benchmark/inference_datasets/op_rna.h5ad',
+    'tf_all': 'resources/grn_benchmark/prior/tf_all.csv',
     'prediction': 'output/grn.h5ad',
     'filtration': 'top-k',
     'max_n_links': 50_000,
@@ -148,12 +148,12 @@ def main(par):
 
 
 if __name__ == '__main__':
-    if True: #TODO:remove this
+    if False: #TODO:remove this
         populate_my_ontology( 
             organisms= ["NCBITaxon:9606"]
         )
     par['checkpoint'] = par['temp_dir'] + '/scprint.ckpt'
-    if True: 
+    if False: 
 
         os.makedirs(par['temp_dir'], exist_ok=True)
         print(f"Downloading checkpoint")