bug in workflow fixed

scripts/run_grn_evaluation.sh

@@ -6,15 +6,15 @@ viash ns build --parallel
 reg_type=ridge
 
 RUN_ID="grn_evaluation_all_${reg_type}"
-# resources_dir="s3://openproblems-data/resources/grn"
-resources_dir="./resources"
+resources_dir="s3://openproblems-data/resources/grn"
+# resources_dir="./resources"
 publish_dir="${resources_dir}/results/${RUN_ID}"
 grn_models_folder="${resources_dir}/grn_models"
 
 subsample=-2
 max_workers=10
 layer=scgen_pearson
-metric_ids="[regression_1]"
+metric_ids="[regression_1, regression_2]"
 
 param_file="./params/${RUN_ID}.yaml"
 
@@ -75,21 +75,21 @@ HERE
 
 }
 
-#Loop through grn_names and layers
-# for grn_name in "${grn_names[@]}"; do
-#   append_entry "$grn_name" 
-# done
+Loop through grn_names and layers
+for grn_name in "${grn_names[@]}"; do
+  append_entry "$grn_name" 
+done
 
 ## controls
-# append_entry_control "negative_control" "" "" "false" "false" "false"
-# append_entry_control "positive_control" "" "" "false" "false" "false"
-append_entry_control "baseline_pearson" "False" "pearson" "false" "false" "false"
-append_entry_control "baseline_dotproduct" "False" "dotproduct" "false" "false" "false"
-append_entry_control "baseline_dotproduct_causal" "True" "dotproduct" "false" "false" "false"
-append_entry_control "baseline_dotproduct_causal_cell_type" "True" "dotproduct" "true" "false" "false"
-append_entry_control "baseline_dotproduct_causal_metacell" "True" "dotproduct" "false" "true" "false"
-append_entry_control "baseline_dotproduct_causal_impute" "True" "dotproduct" "false" "false" "true"
-# append_entry_control "baseline_corr_causal_spearman" "True" "spearman"
+append_entry_control "negative_control" "" "" "false" "false" "false"
+append_entry_control "positive_control" "" "" "false" "false" "false"
+append_entry_control "baseline_pearson" "false" "pearson" "false" "false" "false"
+append_entry_control "baseline_dotproduct" "false" "dotproduct" "false" "false" "false"
+append_entry_control "baseline_dotproduct_causal" "true" "dotproduct" "false" "false" "false"
+append_entry_control "baseline_dotproduct_causal_cell_type" "true" "dotproduct" "true" "false" "false"
+append_entry_control "baseline_dotproduct_causal_metacell" "true" "dotproduct" "false" "true" "false"
+append_entry_control "baseline_dotproduct_causal_impute" "true" "dotproduct" "false" "false" "true"
+append_entry_control "baseline_corr_causal_spearman" "true" "spearman"
 
 
 # Append the remaining output_state and publish_dir to the YAML file
@@ -98,13 +98,13 @@ output_state: "state.yaml"
 publish_dir: "$publish_dir"
 HERE
 
-nextflow run . \
-  -main-script  target/nextflow/workflows/run_grn_evaluation/main.nf \
-  -profile docker \
-  -with-trace \
-  -c src/common/nextflow_helpers/labels_ci.config \
-  -params-file ${param_file}
-subl resources/results/grn_evaluation_all_ridge/scores.yaml
+# nextflow run . \
+#   -main-script  target/nextflow/workflows/run_grn_evaluation/main.nf \
+#   -profile docker \
+#   -with-trace \
+#   -c src/common/nextflow_helpers/labels_ci.config \
+#   -params-file ${param_file}
+# subl resources/results/grn_evaluation_all_ridge/scores.yaml
 
 # ./tw-windows-x86_64.exe launch `
 #     https://github.com/openproblems-bio/task_grn_inference.git `

src/api/comp_method.yaml

@@ -15,7 +15,7 @@ functionality:
       example: resources_test/grn-benchmark/multiomics_rna.h5ad
     - name: --prediction
       __merge__: file_prediction.yaml
-      required: true
+      required: false
       direction: output
       example: resources_test/grn_models/collectri.csv
     - name: --tf_all

src/api/comp_metric.yaml

@@ -59,6 +59,9 @@ functionality:
       direction: input
       required: false
       default: scgen_pearson
+    - name: --max_n_links
+      type: integer
+      default: 50000
 
 
 

src/control_methods/baseline_corr/script.py

@@ -58,7 +58,7 @@ def create_corr_net(X: np.ndarray, groups: np.ndarray, method="pearson"):
     return grn
 print('Read data')
 multiomics_rna = ad.read_h5ad(par["multiomics_rna"])
-# multiomics_rna = multiomics_rna[:,:5000] #TODO: togo
+multiomics_rna = multiomics_rna[:,:5000] #TODO: togo
 
 if par['metacell']:
     print('metacell')

src/metrics/regression_1/main.py

@@ -13,6 +13,12 @@
 import os
 
 
+def select_top_links(net, par):
+    print("Number of links reduced to ", par['max_n_links'])
+    net_sorted = net.reindex(net['weight'].abs().sort_values(ascending=False).index)
+    net = net_sorted.head(par['max_n_links']).reset_index(drop=True)
+    return net
+
 class lightgbm_wrapper:
     def __init__(self, params, max_workers=None):
         self.params =  params
@@ -146,6 +152,8 @@ def regression_1(
             net_sub = net[net.cell_type==cell_type]
         else:
             net_sub = net
+
+        net_sub = select_top_links(net_sub, par) #only top links are considered
 
         prturb_adata_sub = prturb_adata[prturb_adata.obs.cell_type==cell_type,:]
         y_true_sub, y_pred_sub = cross_validation(net_sub, prturb_adata_sub, par)

src/workflows/run_grn_evaluation/main.nf

@@ -32,6 +32,7 @@ workflow run_wf {
       },
       fromState: [
         perturbation_data: "perturbation_data",
+        multiomics_rna: "multiomics_rna",
         layer: "layer",
         tf_all: "tf_all"
       ],
@@ -175,6 +176,7 @@ workflow run_wf {
       },
       fromState: [
         perturbation_data: "perturbation_data",
+        multiomics_rna: "multiomics_rna",
         tf_all: "tf_all",
       ],
       toState: {id, output, state ->