workflow updated-double scglue is removed

runs.ipynb

@@ -170,17 +170,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 1,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Submitted batch job 7851343\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "from src.helper import calculate_scores\n",
     "if False: # consensus: run this after updating grns\n",

src/metrics/regression_2/main.py

@@ -325,7 +325,10 @@ def main(par: Dict[str, Any]) -> pd.DataFrame:
         # Evaluate GRN
         verbose_print(par['verbose'], f'Compute metrics for layer: {layer}', 3)
         verbose_print(par['verbose'], f'Static approach (theta=0):', 3)
-        score_static_min = static_approach(net_matrix, n_features_theta_min, X, groups, gene_names, tf_names, par['reg_type'], n_jobs=par['num_workers'])
+        if (n_features_theta_min!=0).any()==False:
+            score_static_min = np.nan
+        else:
+            score_static_min = static_approach(net_matrix, n_features_theta_min, X, groups, gene_names, tf_names, par['reg_type'], n_jobs=par['num_workers'])
         verbose_print(par['verbose'], f'Static approach (theta=0.5):', 3)
         score_static_median = static_approach(net_matrix, n_features_theta_median, X, groups, gene_names, tf_names, par['reg_type'], n_jobs=par['num_workers'])
         print(f'Static approach (theta=1):', flush=True)

src/metrics/regression_2/script.py

@@ -6,12 +6,12 @@
 
 ## VIASH START
 par = {
-    'evaluation_data': 'resources/grn-benchmark/evaluation_data.h5ad',
+    'evaluation_data': 'resources/evaluation_datasets/op_perturbation.h5ad',
     'layer': 'X_norm',
-    "prediction": "resources/grn_models/celloracle.csv",
+    "prediction": "output/models/collectri.csv",
     'tf_all': 'resources/prior/tf_all.csv',
     "max_n_links": 50000,
-    'consensus': 'resources/prior/consensus-num-regulators.json',
+    'consensus': 'output/models/op_consensus-num-regulators.json',
     'score': 'output/score_regression2.h5ad',
     'reg_type': 'ridge',
     'static_only': True,

src/utils/util.py

@@ -114,24 +114,3 @@ def read_gmt(file_path:str) -> dict[str, list[str]]:
                 'genes': genes
             }
     return gene_sets
-def quantile_transformation(values, one_sided=False, log1p_scale=True):
-    from sklearn.preprocessing import QuantileTransformer
-    if log1p_scale:
-        log_data = np.log1p(values)  # log(x + 1) to avoid log(0)
-    if one_sided:
-        output_distribution = 'uniform'
-    else:
-        output_distribution = 'normal'
-    quantile_transformer = QuantileTransformer(output_distribution=output_distribution)
-    transformed_data = quantile_transformer.fit_transform(log_data.reshape(-1, 1)).reshape(len(log_data))
-    return transformed_data
-def zscore_transformation(values, one_sided=False, log1p_scale=True):
-    if log1p_scale:
-        log_data = np.log1p(values)  # log(x + 1) to avoid log(0)
-    if one_sided:
-        mean = 0
-    else:
-        mean = np.mean(values)
-    std = np.std(values)
-    transformed_data = (log_data-mean)/std
-    return transformed_data

src/workflows/run_benchmark/main.nf

@@ -17,7 +17,6 @@ workflow run_wf {
     grnboost2,
     ppcor,
     scenic,
-    scglue,
     scgpt,
 
     pearson_corr,