gs recovery debugged

Author: janursa

common

@@ -8,6 +8,8 @@ run_grn_inference=false
 run_grn_evaluation=true
 run_download=false
 
+num_workers=20
+
 
 for dataset in "${datasets[@]}"; do
     trace_file="resources/results/$dataset/trace.txt"
@@ -56,7 +58,7 @@ for dataset in "${datasets[@]}"; do
         # fi
 
         echo "Running GRN evaluation for dataset: $dataset"
-        bash scripts/run_grn_evaluation.sh --dataset=$dataset --run_local=$run_local --build_images=false 
+        bash scripts/run_grn_evaluation.sh --dataset=$dataset --run_local=$run_local --build_images=false --num_workers=$num_workers
     fi
 
     if [ "$run_download" = true ]; then

scripts/run_all.sh

@@ -8,8 +8,9 @@ RUN_LOCAL="true"
 RUN_TEST=false
 PREDICTION="none"
 SAVE_DIR="none"
-BUILD_IMAGES=false
+BUILD_IMAGES=true
 reg_type="ridge"
+num_workers=4
 
 
 # Parse arguments
@@ -43,6 +44,10 @@ for arg in "$@"; do
             reg_type="${arg#*=}"
             shift
             ;;
+        --num_workers=*)
+            num_workers="${arg#*=}"
+            shift
+            ;;
         *)
             echo "Unknown argument: $arg"
             exit 1
@@ -63,8 +68,6 @@ if [ -z "${DATASET:-}" ]; then
     exit 1
 fi
 
-
-num_workers=10
 RUN_ID="${DATASET}_evaluation"
 models_folder="${DATASET}/"
 apply_tf=true
@@ -132,13 +135,15 @@ append_entry() {
   else
       layer_=$layer
   fi
+
   cat >> "$param_local" << HERE
   - id: ${grn_name}_${dataset}
     metric_ids: ${metric_ids}
     evaluation_data: ${resources_dir}/grn_benchmark/evaluation_data/${dataset}_bulk.h5ad
     tf_all: ${resources_dir}/grn_benchmark/prior/tf_all.csv
     regulators_consensus: ${resources_dir}/grn_benchmark/prior/regulators_consensus_${dataset}.json
     prediction: ${prediction}
+    num_workers: ${num_workers}
     apply_tf: ${apply_tf}
     reg_type: ${reg_type}
     layer: $layer_
@@ -229,7 +234,7 @@ HERE
     -profile docker \
     -with-trace \
     -params-file ${param_local} \
-    # -c common/nextflow_helpers/labels_ci.config
+    -c common/nextflow_helpers/labels_ci.config
 else
   cat >> "$param_file" << HERE
 output_state: "state.yaml"

scripts/run_grn_evaluation.sh

@@ -17,15 +17,15 @@ set -e
 # aws s3 sync  s3://openproblems-data/resources/grn/results/op resources/results/op  --no-sign-request
 
 # aws s3 sync  s3://openproblems-data/resources/grn/grn_models resources/grn_models --delete 
-# aws s3 sync  resources_test/ s3://openproblems-data/resources_test/grn/  --delete 
 
 
 
 # aws s3 sync   s3://openproblems-data/resources/grn/grn_benchmark resources/grn_benchmark/ --no-sign-request
 # aws s3 sync  resources/grn_benchmark/prior s3://openproblems-data/resources/grn/grn_benchmark/prior --delete
 # aws s3 sync  resources/extended_data/ s3://openproblems-data/resources/grn/extended_data --delete
 # aws s3 sync resources/results/experiment  s3://openproblems-data/resources/grn/results/experiment --delete 
-aws s3 sync resources_test  s3://openproblems-data/resources_test/grn/ --delete 
+# aws s3 sync resources_test  s3://openproblems-data/resources_test/grn/ --delete 
+aws s3 sync  s3://openproblems-data/resources_test/grn/ resources_test --delete 
 # aws s3 sync  resources/grn_benchmark/ground_truth s3://openproblems-data/resources/grn/grn_benchmark/ground_truth 
 # aws s3 sync  resources/grn_benchmark/ s3://openproblems-data/resources/grn/grn_benchmark/ --delete
 

scripts/sync_resources.sh

@@ -15,20 +15,20 @@
     'num_workers': 20,
     'tf_all': 'resources/grn_benchmark/prior/tf_all.csv',    
     'score': 'output/score.h5ad',
-    'pathway_file': '/vol/projects/jnourisa/prior/h.all.v2024.1.Hs.symbols.gmt',
     'fdr_threshold': 0.05,
     'min_pathway_size': 5,
     'max_pathway_size': 500,
     'min_targets': 10,
     'max_targets': 100,  # Top K edges by absolute weight
-    'ulm_baseline_method': 'zero_centered',
-    'output_detailed_metrics': True
+    'ulm_baseline_method': 'zero_centered'
 }
 ## VIASH END
 
+run_local=False
 try:
     sys.path.append(meta["resources_dir"])
 except:
+    run_local=True
     meta = {
     "resources_dir":'src/metrics/experimental/annotation/',
     "util_dir": 'src/utils',
@@ -38,10 +38,11 @@
 from helper import main as main_helper
 from util import format_save_score, parse_args
 
-par = parse_args(par)
+if run_local:
+    par = parse_args(par)
 
 if __name__ == "__main__":
-    
+    print(par)
     output = main_helper(par)
     print(output)
 

src/metrics/gs_recovery/script.py

@@ -19,10 +19,10 @@ METRICS_replogle="regression,ws_distance,tf_recovery,tf_binding,sem,gs_recovery"
 METRICS_adamson="regression,tf_binding,sem,gs_recovery"
 METRICS_norman="regression,ws_distance,tf_binding,sem,gs_recovery"
 METRICS_nakatake="regression,sem,gs_recovery"
-METRICS_op="regression,vc,rc_tf_act,tf_binding,sem,rc,anchor_regression,gs_recovery"
-METRICS_300BCG="regression,vc,rc_tf_act,tf_binding,sem,rc,anchor_regression,gs_recovery"
-METRICS_ibd_uc="regression,vc,tf_binding,sem,anchor_regression,gs_recovery"
-METRICS_ibd_cd="regression,vc,tf_binding,sem,anchor_regression,gs_recovery"
-METRICS_parsebioscience="regression,vc,rc_tf_act,tf_binding,sem,rc,anchor_regression,gs_recovery"
+METRICS_op="gs_recovery"
+METRICS_300BCG="regression,vc,rc_tf_act,tf_binding,sem,gs_recovery"
+METRICS_ibd_uc="regression,vc,tf_binding,sem,gs_recovery"
+METRICS_ibd_cd="regression,vc,tf_binding,sem,gs_recovery"
+METRICS_parsebioscience="regression,vc,rc_tf_act,tf_binding,sem,gs_recovery"
 METRICS_xaira_HEK293T="regression,ws_distance,tf_recovery,tf_binding,sem,gs_recovery"
 METRICS_xaira_HCT116="regression,ws_distance,tf_recovery,tf_binding,sem,gs_recovery"