pheval 0.4.2__tar.gz → 0.4.4__tar.gz

{pheval-0.4.2 → pheval-0.4.4}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.3
 Name: pheval
-Version: 0.4.2
+Version: 0.4.4
 Summary: 
 Author: Yasemin Bridges
 Author-email: y.bridges@qmul.ac.uk

{pheval-0.4.2 → pheval-0.4.4}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "pheval"
-version = "0.4.2"
+version = "0.4.4"
 description = ""
 authors = ["Yasemin Bridges <y.bridges@qmul.ac.uk>",
   "Julius Jacobsen <j.jacobsen@qmul.ac.uk>",

{pheval-0.4.2 → pheval-0.4.4}/src/pheval/analyse/benchmark_db_manager.py

@@ -75,6 +75,7 @@ class BenchmarkDBManager:
                 `False` otherwise.
         """
         list_pattern = re.compile(r"^\[\s*(?:[^\[\],\s]+(?:\s*,\s*[^\[\],\s]+)*)?\s*]$")
+        entity = entity.replace("nan", "None").replace("NaN", "None")
         if list_pattern.match(str(entity)):
             list_representation = ast.literal_eval(entity)
             if isinstance(list_representation, list):

{pheval-0.4.2 → pheval-0.4.4}/src/pheval/analyse/disease_prioritisation_analysis.py

@@ -38,11 +38,16 @@ class AssessDiseasePrioritisation(AssessPrioritisationBase):
         for _i, row in df.iterrows():
             result = (
                 self.conn.execute(
-                    f"SELECT * FROM '{standardised_disease_result_path}' "
-                    f"WHERE contains_entity_function(CAST(COALESCE(disease_identifier, '') AS VARCHAR),"
-                    f" '{row['disease_identifier']}') "
-                    f"OR contains_entity_function(CAST(COALESCE(disease_name, '') AS VARCHAR), "
-                    f"'{row['disease_name']}')"
+                    (
+                        f"SELECT * FROM '{standardised_disease_result_path}' "
+                        f"WHERE contains_entity_function(CAST(COALESCE(disease_identifier, '') AS VARCHAR),"
+                        f" '{row['disease_identifier']}') "
+                        f"OR contains_entity_function(CAST(COALESCE(disease_name, '') AS VARCHAR), "
+                        f"'{row['disease_name']}')"
+                    )
+                    if standardised_disease_result_path.exists()
+                    and standardised_disease_result_path.stat().st_size > 0
+                    else "SELECT NULL WHERE FALSE"
                 )
                 .fetchdf()
                 .to_dict(orient="records")
@@ -56,9 +61,15 @@ class AssessDiseasePrioritisation(AssessPrioritisationBase):
                     f'UPDATE {self.table_name} SET "{self.column}" = ? WHERE identifier = ?',
                     (disease_match, primary_key),
                 )
+            elif len(result) == 0:
+                relevant_ranks.append(0)
         binary_classification_stats.add_classification(
-            self.db_connection.parse_table_into_dataclass(
-                str(standardised_disease_result_path), RankedPhEvalDiseaseResult
+            (
+                self.db_connection.parse_table_into_dataclass(
+                    str(standardised_disease_result_path), RankedPhEvalDiseaseResult
+                )
+                if standardised_disease_result_path.exists()
+                else []
             ),
             relevant_ranks,
         )

{pheval-0.4.2 → pheval-0.4.4}/src/pheval/analyse/gene_prioritisation_analysis.py

@@ -36,11 +36,16 @@ class AssessGenePrioritisation(AssessPrioritisationBase):
         for _i, row in df.iterrows():
             result = (
                 self.conn.execute(
-                    f"SELECT * FROM '{standardised_gene_result_path}' "
-                    f"WHERE contains_entity_function(CAST(COALESCE(gene_identifier, '') AS VARCHAR),"
-                    f" '{row['gene_identifier']}') "
-                    f"OR contains_entity_function(CAST(COALESCE(gene_symbol, '') AS VARCHAR), "
-                    f"'{row['gene_symbol']}')"
+                    (
+                        f"SELECT * FROM '{standardised_gene_result_path}' "
+                        f"WHERE contains_entity_function(CAST(COALESCE(gene_identifier, '') AS VARCHAR), "
+                        f"'{row['gene_identifier']}') "
+                        f"OR contains_entity_function(CAST(COALESCE(gene_symbol, '') AS VARCHAR), "
+                        f"'{row['gene_symbol']}')"
+                    )
+                    if standardised_gene_result_path.exists()
+                    and standardised_gene_result_path.stat().st_size > 0
+                    else "SELECT NULL WHERE FALSE"
                 )
                 .fetchdf()
                 .to_dict(orient="records")
@@ -53,9 +58,15 @@ class AssessGenePrioritisation(AssessPrioritisationBase):
                     f'UPDATE {self.table_name} SET "{self.column}" = ? WHERE identifier = ?',
                     (gene_match, primary_key),
                 )
+            if not result:
+                relevant_ranks.append(0)
         binary_classification_stats.add_classification(
-            self.db_connection.parse_table_into_dataclass(
-                str(standardised_gene_result_path), RankedPhEvalGeneResult
+            (
+                self.db_connection.parse_table_into_dataclass(
+                    str(standardised_gene_result_path), RankedPhEvalGeneResult
+                )
+                if standardised_gene_result_path.exists()
+                else []
             ),
             relevant_ranks,
         )

{pheval-0.4.2 → pheval-0.4.4}/src/pheval/analyse/generate_plots.py

@@ -2,6 +2,7 @@ from pathlib import Path
 from typing import List
 
 import matplotlib
+import numpy as np
 import pandas as pd
 import seaborn as sns
 from matplotlib import pyplot as plt
@@ -368,9 +369,16 @@ class PlotGenerator:
         """
         plt.clf()
         for i, benchmark_result in enumerate(benchmarking_results):
+            y_score = np.array(benchmark_result.binary_classification_stats.scores)
+            y_score = np.nan_to_num(
+                y_score,
+                nan=0.0,
+                posinf=max(y_score[np.isfinite(y_score)]),
+                neginf=min(y_score[np.isfinite(y_score)]),
+            )
             fpr, tpr, thresh = roc_curve(
                 benchmark_result.binary_classification_stats.labels,
-                benchmark_result.binary_classification_stats.scores,
+                y_score,
                 pos_label=1,
             )
             roc_auc = auc(fpr, tpr)
@@ -411,9 +419,16 @@ class PlotGenerator:
         plt.clf()
         plt.figure()
         for i, benchmark_result in enumerate(benchmarking_results):
+            y_score = np.array(benchmark_result.binary_classification_stats.scores)
+            y_score = np.nan_to_num(
+                y_score,
+                nan=0.0,
+                posinf=max(y_score[np.isfinite(y_score)]),
+                neginf=min(y_score[np.isfinite(y_score)]),
+            )
             precision, recall, thresh = precision_recall_curve(
                 benchmark_result.binary_classification_stats.labels,
-                benchmark_result.binary_classification_stats.scores,
+                y_score,
             )
             precision_recall_auc = auc(recall, precision)
             plt.plot(

{pheval-0.4.2 → pheval-0.4.4}/src/pheval/analyse/variant_prioritisation_analysis.py

@@ -44,12 +44,16 @@ class AssessVariantPrioritisation(AssessPrioritisationBase):
             )
             result = (
                 self.conn.execute(
-                    f"SELECT * FROM '{standardised_variant_result_path}' "
-                    f"WHERE "
-                    f"chromosome == '{causative_variant.chrom}' AND "
-                    f"start == {causative_variant.pos} AND "
-                    f"ref == '{causative_variant.ref}' AND "
-                    f"alt == '{causative_variant.alt}'"
+                    (
+                        f"SELECT * FROM '{standardised_variant_result_path}' "
+                        f"WHERE "
+                        f"chromosome == '{causative_variant.chrom}' AND "
+                        f"start == {causative_variant.pos} AND "
+                        f"ref == '{causative_variant.ref}' AND "
+                        f"alt == '{causative_variant.alt}'"
+                    )
+                    if standardised_variant_result_path.exists()
+                    else "SELECT NULL WHERE FALSE"
                 )
                 .fetchdf()
                 .to_dict(orient="records")
@@ -66,10 +70,15 @@ class AssessVariantPrioritisation(AssessPrioritisationBase):
                     f'UPDATE {self.table_name} SET "{self.column}" = ? WHERE identifier = ?',
                     (variant_match, primary_key),
                 )
-
+            elif len(result) == 0:
+                relevant_ranks.append(0)
         binary_classification_stats.add_classification(
-            self.db_connection.parse_table_into_dataclass(
-                str(standardised_variant_result_path), RankedPhEvalVariantResult
+            (
+                self.db_connection.parse_table_into_dataclass(
+                    str(standardised_variant_result_path), RankedPhEvalVariantResult
+                )
+                if standardised_variant_result_path.exists()
+                else []
             ),
             relevant_ranks,
         )

{pheval-0.4.2 → pheval-0.4.4}/src/pheval/prepare/prepare_corpus.py

@@ -90,6 +90,10 @@ def prepare_corpus(
             )
         else:
             # if not updating phenopacket gene identifiers then copy phenopacket as is to output directory
-            shutil.copy(
-                phenopacket_path, output_dir.joinpath(f"phenopackets/{phenopacket_path.name}")
+            (
+                shutil.copy(
+                    phenopacket_path, output_dir.joinpath(f"phenopackets/{phenopacket_path.name}")
+                )
+                if phenopacket_path != output_dir.joinpath(f"phenopackets/{phenopacket_path.name}")
+                else None
             )