consensus added to evluation

scripts/run_grn_evaluation_tw.sh

@@ -29,51 +29,35 @@ cat > $param_file << HERE
 param_list:
 HERE
 
-# Nested loops to iterate over grn_names and layers
-for grn_name in "${grn_names[@]}"; do
-  for layer in "${layers[@]}"; do
-    cat >> $param_file << HERE
-  - id: ${layer}_${grn_name}
+append_entry() {
+  cat >> $param_file << HERE
+  - id: ${layer}_${1}
     perturbation_data: ${resources_dir}/grn-benchmark/perturbation_data.h5ad
     layer: ${layer}
-    prediction: ${resources_dir}/grn_models/${grn_name}.csv 
     reg_type: $reg_type
-    method_id: $grn_name
+    method_id: $1
     subsample: $subsample
     max_workers: $max_workers
-
+    consensus: ${resources_dir}/prior/consensus.json
+    ${2:+tf_all: ${resources_dir}/prior/tf_all.csv}
+    ${3:+prediction: ${resources_dir}/grn_models/$1.csv}
 HERE
+}
+# Loop through grn_names and layers
+for grn_name in "${grn_names[@]}"; do
+  for layer in "${layers[@]}"; do
+    append_entry "$grn_name" "" "true"
   done
 done
 
-
-# append negative control
+# Append negative control
 grn_name="negative_control"
-cat >> $param_file << HERE
-  - id: ${layer}_${grn_name}
-    perturbation_data: ${resources_dir}/grn-benchmark/perturbation_data.h5ad
-    layer: ${layer}
-    reg_type: $reg_type
-    method_id: $grn_name
-    subsample: $subsample
-    max_workers: $max_workers
-
-HERE
+append_entry "$grn_name"
 
-# append the positive controls
+# Append positive controls
 grn_name="positive_control"
 for layer in "${layers[@]}"; do
-  cat >> $param_file << HERE
-  - id: ${layer}_${grn_name}
-    perturbation_data: ${resources_dir}/grn-benchmark/perturbation_data.h5ad
-    tf_all: ${resources_dir}/prior/tf_all.csv
-    layer: ${layer}
-    reg_type: $reg_type
-    method_id: $grn_name
-    subsample: $subsample
-    max_workers: $max_workers
-
-HERE
+  append_entry "$grn_name" "true"
 done
 
 # Append the remaining output_state and publish_dir to the YAML file

src/methods/multi_omics/celloracle/main.py

@@ -65,7 +65,10 @@ def preprocess_rna(par) -> None:
                                                 log=False)
 
     # Subset the genes
-    adata = adata[:, filter_result.gene_subset]
+    if False: #only hvgs
+        adata = adata[:, filter_result.gene_subset]
+    else:
+        pass
 
     # Renormalize after filtering
     sc.pp.normalize_per_cell(adata)

src/methods/multi_omics/scglue_ns/run.sh

@@ -24,3 +24,4 @@ param_list:
 output_state: "state.yaml"
 publish_dir: "$publish_dir"
 HERE
+

src/metrics/regression_2/config.vsh.yaml

@@ -15,7 +15,7 @@ functionality:
       type: file
       direction: input
       must_exist: true
-      default: 'resources/grn-benchmark/consensus-num-regulators.json'
+      default: 'resources/prior/consensus-num-regulators.json'
 platforms:
   - type: docker
     image: ghcr.io/openproblems-bio/base_python:1.0.4

src/workflows/run_grn_evaluation/config.vsh.yaml

@@ -41,6 +41,11 @@ functionality:
           required: false
           direction: input
           default: resources/prior/tf_all.csv
+        - name: --consensus
+          type: file
+          required: false
+          direction: input
+          default: resources/prior/consensus.json
 
     - name: Outputs
       arguments:

src/workflows/run_grn_evaluation/main.nf

@@ -69,7 +69,8 @@ workflow run_wf {
         subsample: "subsample",
         reg_type: "reg_type",
         method_id: "method_id",
-        max_workers: "max_workers"
+        max_workers: "max_workers",
+        consensus: "consensus"
       ],
       // use 'toState' to publish that component's outputs to the overall state
       toState: { id, output, state, comp ->