add counts after each step

EBIvariation · Jul 12, 2023 · aa24d59 · aa24d59
1 parent 29fbd00
commit aa24d59
Show file tree

Hide file tree

Showing 2 changed files with 17 additions and 9 deletions.
diff --git a/opentargets_pharmgkb/counts.py b/opentargets_pharmgkb/counts.py
@@ -5,8 +5,10 @@ def __init__(self):
         # Input counts (before annotation and exploding by drug, etc.)
         self.clinical_annotations = 0
         self.with_rs = 0
-        # Counts after exploding by allele
-        self.allele_annotations = 0
+        # Counts after exploding by each attribute
+        self.exploded_alleles = 0
+        self.exploded_drugs = 0
+        self.exploded_phenotypes = 0
         # Output counts (after annotation and exploding)
         self.evidence_strings = 0
         self.with_chebi = 0
@@ -22,7 +24,9 @@ def __init__(self):
     def report(self):
         report_str = f'\nTotal clinical annotations: {self.clinical_annotations}\n'
         report_str += f'\tWith RS: {self.with_rs}\n'
-        report_str += f'Total annotations by allele: {self.allele_annotations}\n'
+        report_str += f'Exploded by allele: {self.exploded_alleles}\n'
+        report_str += f'Exploded by drug: {self.exploded_drugs}\n'
+        report_str += f'Exploded by phenotype: {self.exploded_phenotypes}\n'
         report_str += f'Total evidence strings: {self.evidence_strings}\n'
         report_str += f'\tWith CHEBI: {self.with_chebi}\n'
         report_str += f'\tWith EFO phenotype: {self.with_efo}\n'

diff --git a/opentargets_pharmgkb/evidence_generation.py b/opentargets_pharmgkb/evidence_generation.py
@@ -45,16 +45,20 @@ def pipeline(data_dir, created_date, output_path):
 
     # Main processing
     merged_with_alleles_table = pd.merge(rs_only_table, clinical_alleles_table, on=ID_COL_NAME, how='left')
-    counts.allele_annotations = len(merged_with_alleles_table)
-    coordinates_table = get_vcf_coordinates(merged_with_alleles_table)
-    consequences_table = get_functional_consequences(coordinates_table)
-    # mapped_genes = explode_and_map_genes(consequences_table)
-    mapped_drugs = explode_and_map_drugs(consequences_table, drugs_table)
+    counts.exploded_alleles = len(merged_with_alleles_table)
+
+    mapped_drugs = explode_and_map_drugs(merged_with_alleles_table, drugs_table)
+    counts.exploded_drugs = len(mapped_drugs)
+
     mapped_phenotypes = explode_and_map_phenotypes(mapped_drugs)
+    counts.exploded_phenotypes = len(mapped_phenotypes)
+
+    coordinates_table = get_vcf_coordinates(mapped_phenotypes)
+    consequences_table = get_functional_consequences(coordinates_table)
 
     # Add clinical evidence with PMIDs
     pmid_evidence = clinical_evidence_table[clinical_evidence_table['PMID'].notna()]
-    evidence_table = pd.merge(mapped_phenotypes, pmid_evidence.groupby(by=ID_COL_NAME).aggregate(
+    evidence_table = pd.merge(consequences_table, pmid_evidence.groupby(by=ID_COL_NAME).aggregate(
         publications=('PMID', list)), on=ID_COL_NAME)
 
     # Gather output counts