feat: add input nodes as evaluation baseline

Snakefile

@@ -430,13 +430,17 @@ def get_gold_standard_pickle_file(wildcards):
     gs = parts[1]
     return SEP.join([out_dir, f'gs-{gs}-merged.pickle'])
 
+# Return the dataset pickle file for a specific dataset
+def get_dataset_pickle_file(wildcards):
+    dataset_label = get_dataset_label(wildcards)
+    return SEP.join([out_dir, f'dataset-{dataset_label}-merged.pickle'])
+
 # Returns the dataset corresponding to the gold standard pair
 def get_dataset_label(wildcards):
     parts = wildcards.dataset_gold_standard_pair.split('-')
     dataset = parts[0]
     return dataset
 
-
 # Returns all pathways for a specific dataset
 def collect_pathways_per_dataset(wildcards):
     dataset_label = get_dataset_label(wildcards)
@@ -446,14 +450,16 @@ def collect_pathways_per_dataset(wildcards):
 rule evaluation_pr_per_pathways:
     input: 
         node_gold_standard_file = get_gold_standard_pickle_file,
+        dataset_file = get_dataset_pickle_file,
         pathways = collect_pathways_per_dataset
     output: 
         node_pr_file = SEP.join([out_dir, '{dataset_gold_standard_pair}-eval', "pr-per-pathway-nodes.txt"]),
         node_pr_png = SEP.join([out_dir, '{dataset_gold_standard_pair}-eval', 'pr-per-pathway-nodes.png']),
     run:
         node_table = Evaluation.from_file(input.node_gold_standard_file).node_table
+        input_nodes = Dataset.from_file(input.dataset_file).get_input_nodes()
         pr_df = Evaluation.node_precision_and_recall(input.pathways, node_table)
-        Evaluation.precision_and_recall_per_pathway(pr_df, output.node_pr_file, output.node_pr_png)
+        Evaluation.precision_and_recall_per_pathway(pr_df, input_nodes, node_table, output.node_pr_file, output.node_pr_png)
 
 # Returns all pathways for a specific algorithm and dataset
 def collect_pathways_per_algo_per_dataset(wildcards):
@@ -465,14 +471,16 @@ def collect_pathways_per_algo_per_dataset(wildcards):
 rule evaluation_per_algo_pr_per_pathways:
     input: 
         node_gold_standard_file = get_gold_standard_pickle_file,
+        dataset_file = get_dataset_pickle_file,
         pathways =  collect_pathways_per_algo_per_dataset,
     output: 
         node_pr_file = SEP.join([out_dir, '{dataset_gold_standard_pair}-eval', "pr-per-pathway-for-{algorithm}-nodes.txt"]),
         node_pr_png = SEP.join([out_dir, '{dataset_gold_standard_pair}-eval', 'pr-per-pathway-for-{algorithm}-nodes.png']),
     run:
         node_table = Evaluation.from_file(input.node_gold_standard_file).node_table
+        input_nodes = Dataset.from_file(input.dataset_file).get_input_nodes()
         pr_df = Evaluation.node_precision_and_recall(input.pathways, node_table)
-        Evaluation.precision_and_recall_per_pathway(pr_df, output.node_pr_file, output.node_pr_png, include_aggregate_algo_eval)
+        Evaluation.precision_and_recall_per_pathway(pr_df, input_nodes, node_table, output.node_pr_file, output.node_pr_png, include_aggregate_algo_eval)
 
 # Return pathway summary file per dataset
 def collect_summary_statistics_per_dataset(wildcards):
@@ -490,16 +498,18 @@ def collect_pca_coordinates_per_dataset(wildcards):
 rule evaluation_pca_chosen:
     input: 
         node_gold_standard_file = get_gold_standard_pickle_file,
+        dataset_file = get_dataset_pickle_file,
         pca_coordinates_file = collect_pca_coordinates_per_dataset,
         pathway_summary_file = collect_summary_statistics_per_dataset
     output: 
         node_pca_chosen_pr_file = SEP.join([out_dir, '{dataset_gold_standard_pair}-eval', 'pr-pca-chosen-pathway-nodes.txt']),
         node_pca_chosen_pr_png = SEP.join([out_dir, '{dataset_gold_standard_pair}-eval', 'pr-pca-chosen-pathway-nodes.png']),
     run:
         node_table = Evaluation.from_file(input.node_gold_standard_file).node_table
+        input_nodes = Dataset.from_file(input.dataset_file).get_input_nodes()
         pca_chosen_pathway = Evaluation.pca_chosen_pathway(input.pca_coordinates_file, input.pathway_summary_file, out_dir)
         pr_df = Evaluation.node_precision_and_recall(pca_chosen_pathway, node_table)
-        Evaluation.precision_and_recall_pca_chosen_pathway(pr_df, output.node_pca_chosen_pr_file, output.node_pca_chosen_pr_png)
+        Evaluation.precision_and_recall_pca_chosen_pathway(pr_df, input_nodes, node_table, output.node_pca_chosen_pr_file, output.node_pca_chosen_pr_png)
 
 # Returns pca coordinates for a specific algorithm and dataset
 def collect_pca_coordinates_per_algo_per_dataset(wildcards):
@@ -511,21 +521,18 @@ def collect_pca_coordinates_per_algo_per_dataset(wildcards):
 rule evaluation_per_algo_pca_chosen:
     input: 
         node_gold_standard_file = get_gold_standard_pickle_file,
+        dataset_file = get_dataset_pickle_file,
         pca_coordinates_file = collect_pca_coordinates_per_algo_per_dataset,
         pathway_summary_file = collect_summary_statistics_per_dataset
     output: 
         node_pca_chosen_pr_file = SEP.join([out_dir, '{dataset_gold_standard_pair}-eval', 'pr-pca-chosen-pathway-per-algorithm-nodes.txt']),
         node_pca_chosen_pr_png = SEP.join([out_dir, '{dataset_gold_standard_pair}-eval', 'pr-pca-chosen-pathway-per-algorithm-nodes.png']),
     run:
         node_table = Evaluation.from_file(input.node_gold_standard_file).node_table
+        input_nodes = Dataset.from_file(input.dataset_file).get_input_nodes()
         pca_chosen_pathways = Evaluation.pca_chosen_pathway(input.pca_coordinates_file, input.pathway_summary_file, out_dir)
         pr_df = Evaluation.node_precision_and_recall(pca_chosen_pathways, node_table)
-        Evaluation.precision_and_recall_pca_chosen_pathway(pr_df, output.node_pca_chosen_pr_file, output.node_pca_chosen_pr_png, include_aggregate_algo_eval)
-
-# Return the dataset pickle file for a specific dataset
-def get_dataset_pickle_file(wildcards):
-    dataset_label = get_dataset_label(wildcards)
-    return SEP.join([out_dir, f'dataset-{dataset_label}-merged.pickle'])
+        Evaluation.precision_and_recall_pca_chosen_pathway(pr_df, input_nodes, node_table, output.node_pca_chosen_pr_file, output.node_pca_chosen_pr_png, include_aggregate_algo_eval)
 
 # Returns ensemble file for each dataset
 def collect_ensemble_per_dataset(wildcards):
@@ -543,8 +550,9 @@ rule evaluation_ensemble_pr_curve:
         node_pr_curve_file = SEP.join([out_dir, '{dataset_gold_standard_pair}-eval', 'pr-curve-ensemble-nodes.txt']),
     run:
         node_table = Evaluation.from_file(input.node_gold_standard_file).node_table
+        input_nodes = Dataset.from_file(input.dataset_file).get_input_nodes()
         node_ensemble_dict = Evaluation.edge_frequency_node_ensemble(node_table, input.ensemble_file, input.dataset_file)
-        Evaluation.precision_recall_curve_node_ensemble(node_ensemble_dict, node_table, output.node_pr_curve_png, output.node_pr_curve_file)
+        Evaluation.precision_recall_curve_node_ensemble(node_ensemble_dict, node_table, input_nodes, output.node_pr_curve_png, output.node_pr_curve_file)
 
 # Returns list of algorithm specific ensemble files per dataset
 def collect_ensemble_per_algo_per_dataset(wildcards):
@@ -562,8 +570,9 @@ rule evaluation_per_algo_ensemble_pr_curve:
         node_pr_curve_file = SEP.join([out_dir, '{dataset_gold_standard_pair}-eval', 'pr-curve-ensemble-nodes-per-algorithm-nodes.txt']),
     run:
         node_table = Evaluation.from_file(input.node_gold_standard_file).node_table
-        node_ensembles_dict = Evaluation.edge_frequency_node_ensemble(node_table, input.ensemble_files, input.dataset_file)
-        Evaluation.precision_recall_curve_node_ensemble(node_ensembles_dict, node_table, output.node_pr_curve_png, output.node_pr_curve_file, include_aggregate_algo_eval)
+        input_nodes = Dataset.from_file(input.dataset_file).get_input_nodes()
+        node_ensembles_dict= Evaluation.edge_frequency_node_ensemble(node_table, input.ensemble_files, input.dataset_file)
+        Evaluation.precision_recall_curve_node_ensemble(node_ensembles_dict, node_table, input_nodes, output.node_pr_curve_png, output.node_pr_curve_file, include_aggregate_algo_eval)
 
 rule evaluation_edge_dummy:
     input: 

spras/dataset.py

@@ -226,6 +226,15 @@ def contains_node_columns(self, col_names: list[str] | str):
                     return False
                 return True
 
+    def get_input_nodes(self) -> pd.DataFrame:
+        """
+        Returns: a table listing the input nodes considered as starting points for pathway reconstruction algorithms,
+        restricted to nodes that have at least one of the specified attributes.
+        """
+        input_node_columns = ["sources", "targets", "prize", "active"] # TODO: do we want to add dummy nodes?
+        input_nodes = Dataset.get_node_columns(self, col_names = input_node_columns)
+        return input_nodes
+
     def get_other_files(self):
         return self.other_files.copy()