pheval 0.3.8__py3-none-any.whl → 0.4.0__py3-none-any.whl

pheval/analyse/analysis.py

@@ -1,193 +1,104 @@
-from collections import defaultdict
-from pathlib import Path
-from typing import List
-
 from pheval.analyse.benchmark_generator import (
     BenchmarkRunOutputGenerator,
     DiseaseBenchmarkRunOutputGenerator,
     GeneBenchmarkRunOutputGenerator,
     VariantBenchmarkRunOutputGenerator,
 )
-from pheval.analyse.generate_summary_outputs import (
-    generate_benchmark_comparison_output,
-    generate_benchmark_output,
-)
+from pheval.analyse.generate_summary_outputs import generate_benchmark_comparison_output
+from pheval.analyse.parse_corpus import CorpusParser
 from pheval.analyse.rank_stats import RankStatsWriter
-from pheval.analyse.run_data_parser import TrackInputOutputDirectories
-
-
-def _run_benchmark(
-    results_dir_and_input: TrackInputOutputDirectories,
-    score_order: str,
-    output_prefix: str,
-    threshold: float,
-    plot_type: str,
-    benchmark_generator: BenchmarkRunOutputGenerator,
-) -> None:
-    """Run a benchmark on a result directory.
-
-    Args:
-        results_dir_and_input (TrackInputOutputDirectories): Input and output directories for tracking results.
-        score_order (str): The order in which scores are arranged, this can be either ascending or descending.
-        output_prefix (str): Prefix for the benchmark output file names.
-        threshold (float): The threshold for benchmark evaluation.
-        plot_type (str): Type of plot for benchmark visualisation.
-        benchmark_generator (BenchmarkRunOutputGenerator): Generator for benchmark run output.
-    """
-    stats_writer = RankStatsWriter(
-        Path(output_prefix + benchmark_generator.stats_comparison_file_suffix)
-    )
-    rank_comparison = defaultdict(dict)
-    benchmark_result = benchmark_generator.generate_benchmark_run_results(
-        results_dir_and_input, score_order, threshold, rank_comparison
-    )
-    stats_writer.write_row(
-        results_dir_and_input.results_dir,
-        benchmark_result.rank_stats,
-        benchmark_result.binary_classification_stats,
-    )
-    generate_benchmark_output(benchmark_result, plot_type, benchmark_generator)
-    stats_writer.close()
-
-
-def benchmark_directory(
-    results_dir_and_input: TrackInputOutputDirectories,
-    score_order: str,
-    output_prefix: str,
-    threshold: float,
-    gene_analysis: bool,
-    variant_analysis: bool,
-    disease_analysis: bool,
-    plot_type: str,
-) -> None:
-    """
-    Benchmark prioritisation performance for a single run.
-
-    Args:
-        results_dir_and_input (TrackInputOutputDirectories): Input and output directories for tracking results.
-        score_order (str): The order in which scores are arranged, this can be either ascending or descending.
-        output_prefix (str): Prefix for the benchmark output file names.
-        threshold (float): The threshold for benchmark evaluation.
-        gene_analysis (bool): Boolean flag indicating whether to benchmark gene results.
-        variant_analysis (bool): Boolean flag indicating whether to benchmark variant results.
-        disease_analysis (bool): Boolean flag indicating whether to benchmark disease results.
-        plot_type (str): Type of plot for benchmark visualisation.
-    """
-    if gene_analysis:
-        _run_benchmark(
-            results_dir_and_input=results_dir_and_input,
-            score_order=score_order,
-            output_prefix=output_prefix,
-            threshold=threshold,
-            plot_type=plot_type,
-            benchmark_generator=GeneBenchmarkRunOutputGenerator(),
-        )
-    if variant_analysis:
-        _run_benchmark(
-            results_dir_and_input=results_dir_and_input,
-            score_order=score_order,
-            output_prefix=output_prefix,
-            threshold=threshold,
-            plot_type=plot_type,
-            benchmark_generator=VariantBenchmarkRunOutputGenerator(),
-        )
-    if disease_analysis:
-        _run_benchmark(
-            results_dir_and_input=results_dir_and_input,
-            score_order=score_order,
-            output_prefix=output_prefix,
-            threshold=threshold,
-            plot_type=plot_type,
-            benchmark_generator=DiseaseBenchmarkRunOutputGenerator(),
-        )
+from pheval.analyse.run_data_parser import Config
 
 
 def _run_benchmark_comparison(
-    results_directories: List[TrackInputOutputDirectories],
-    score_order: str,
-    output_prefix: str,
-    threshold: float,
-    plot_type: str,
+    run_config: Config,
     benchmark_generator: BenchmarkRunOutputGenerator,
 ) -> None:
     """
     Run a benchmark on several result directories.
 
     Args:
-        results_directories (List[TrackInputOutputDirectories]): List of input and output directories
+        run_config (List[TrackInputOutputDirectories]): List of input and output directories
             for tracking results across multiple directories.
-        score_order (str): The order in which scores are arranged, this can be either ascending or descending.
-        output_prefix (str): Prefix for the benchmark output file names.
-        threshold (float): The threshold for benchmark evaluation.
-        plot_type (str): Type of plot for benchmark visualisation.
         benchmark_generator (BenchmarkRunOutputGenerator): Generator for benchmark run output.
     """
     stats_writer = RankStatsWriter(
-        Path(output_prefix + benchmark_generator.stats_comparison_file_suffix)
+        run_config.benchmark_name, benchmark_generator.stats_comparison_file
     )
+    unique_test_corpora_directories = set([result.phenopacket_dir for result in run_config.runs])
+    [
+        CorpusParser(run_config.benchmark_name, test_corpora_directory).parse_corpus(
+            benchmark_generator
+        )
+        for test_corpora_directory in unique_test_corpora_directories
+    ]
     benchmarking_results = []
-    for results_dir_and_input in results_directories:
-        rank_comparison = defaultdict(dict)
+    for run in run_config.runs:
         benchmark_result = benchmark_generator.generate_benchmark_run_results(
-            results_dir_and_input, score_order, threshold, rank_comparison
+            run_config.benchmark_name, run, run.score_order, run.threshold
         )
-        stats_writer.write_row(
-            results_dir_and_input.results_dir,
+        stats_writer.add_statistics_entry(
+            run.run_identifier,
             benchmark_result.rank_stats,
             benchmark_result.binary_classification_stats,
         )
         benchmarking_results.append(benchmark_result)
-    generate_benchmark_comparison_output(benchmarking_results, plot_type, benchmark_generator)
-    stats_writer.close()
+    run_identifiers = [run.run_identifier for run in run_config.runs]
+    [
+        generate_benchmark_comparison_output(
+            run_config.benchmark_name,
+            benchmarking_results,
+            run_identifiers,
+            benchmark_generator,
+            f"{unique_test_corpora_directory.parents[0].name}_"
+            f"{benchmark_generator.prioritisation_type_string}",
+        )
+        for unique_test_corpora_directory in unique_test_corpora_directories
+    ]
 
 
 def benchmark_run_comparisons(
-    results_directories: List[TrackInputOutputDirectories],
-    score_order: str,
-    output_prefix: str,
-    threshold: float,
-    gene_analysis: bool,
-    variant_analysis: bool,
-    disease_analysis: bool,
-    plot_type: str,
+    run_config: Config,
 ) -> None:
     """
     Benchmark prioritisation performance for several runs.
 
     Args:
-        results_directories (List[TrackInputOutputDirectories]): Input and output directories for tracking results.
-        score_order (str): The order in which scores are arranged, this can be either ascending or descending.
-        output_prefix (str): Prefix for the benchmark output file names.
-        threshold (float): The threshold for benchmark evaluation.
-        gene_analysis (bool): Boolean flag indicating whether to benchmark gene results.
-        variant_analysis (bool): Boolean flag indicating whether to benchmark variant results.
-        disease_analysis (bool): Boolean flag indicating whether to benchmark disease results.
-        plot_type (str): Type of plot for benchmark visualisation.
+        run_config (Config): Run configurations.
     """
-    if gene_analysis:
+    gene_analysis_runs = Config(
+        benchmark_name=run_config.benchmark_name,
+        runs=[run for run in run_config.runs if run.gene_analysis],
+        plot_customisation=run_config.plot_customisation,
+    )
+    variant_analysis_runs = Config(
+        benchmark_name=run_config.benchmark_name,
+        runs=[run for run in run_config.runs if run.variant_analysis],
+        plot_customisation=run_config.plot_customisation,
+    )
+    disease_analysis_runs = Config(
+        benchmark_name=run_config.benchmark_name,
+        runs=[run for run in run_config.runs if run.disease_analysis],
+        plot_customisation=run_config.plot_customisation,
+    )
+    if gene_analysis_runs.runs:
         _run_benchmark_comparison(
-            results_directories=results_directories,
-            score_order=score_order,
-            output_prefix=output_prefix,
-            threshold=threshold,
-            plot_type=plot_type,
-            benchmark_generator=GeneBenchmarkRunOutputGenerator(),
+            run_config=gene_analysis_runs,
+            benchmark_generator=GeneBenchmarkRunOutputGenerator(
+                plot_customisation=gene_analysis_runs.plot_customisation.gene_plots
+            ),
         )
-    if variant_analysis:
+    if variant_analysis_runs.runs:
         _run_benchmark_comparison(
-            results_directories=results_directories,
-            score_order=score_order,
-            output_prefix=output_prefix,
-            threshold=threshold,
-            plot_type=plot_type,
-            benchmark_generator=VariantBenchmarkRunOutputGenerator(),
+            run_config=variant_analysis_runs,
+            benchmark_generator=VariantBenchmarkRunOutputGenerator(
+                plot_customisation=variant_analysis_runs.plot_customisation.variant_plots
+            ),
         )
-    if disease_analysis:
+    if disease_analysis_runs.runs:
         _run_benchmark_comparison(
-            results_directories=results_directories,
-            score_order=score_order,
-            output_prefix=output_prefix,
-            threshold=threshold,
-            plot_type=plot_type,
-            benchmark_generator=DiseaseBenchmarkRunOutputGenerator(),
+            run_config=disease_analysis_runs,
+            benchmark_generator=DiseaseBenchmarkRunOutputGenerator(
+                plot_customisation=disease_analysis_runs.plot_customisation.disease_plots
+            ),
         )

pheval/analyse/assess_prioritisation_base.py

@@ -0,0 +1,108 @@
+from typing import Union
+
+from pheval.analyse.benchmark_db_manager import BenchmarkDBManager
+from pheval.post_processing.post_processing import (
+    RankedPhEvalDiseaseResult,
+    RankedPhEvalGeneResult,
+    RankedPhEvalVariantResult,
+)
+
+
+class AssessPrioritisationBase:
+    def __init__(
+        self,
+        db_connection: BenchmarkDBManager,
+        table_name: str,
+        column: str,
+        threshold: float,
+        score_order: str,
+    ):
+        """
+        Initialise AssessPrioritisationBase class
+
+        Args:
+            db_connection (BenchmarkDBManager): DB connection.
+            table_name (str): Table name.
+            column (str): Column name.
+            threshold (float): Threshold for scores
+            score_order (str): Score order for results, either ascending or descending
+
+        """
+        self.threshold = threshold
+        self.score_order = score_order
+        self.db_connection = db_connection
+        self.conn = db_connection.conn
+        self.column = column
+        self.table_name = table_name
+        db_connection.add_column_integer_default(
+            table_name=table_name, column=self.column, default=0
+        )
+
+    def _assess_with_threshold_ascending_order(
+        self,
+        result_entry: Union[
+            RankedPhEvalGeneResult, RankedPhEvalDiseaseResult, RankedPhEvalVariantResult
+        ],
+    ) -> int:
+        """
+        Record the prioritisation rank if it meets the ascending order threshold.
+
+
+        Args:
+            result_entry (Union[RankedPhEvalGeneResult, RankedPhEvalDiseaseResult, RankedPhEvalVariantResult]):
+                Ranked PhEval result entry
+
+        Returns:
+            int: Recorded prioritisation rank
+        """
+        if float(self.threshold) > float(result_entry.score):
+            return result_entry.rank
+        else:
+            return 0
+
+    def _assess_with_threshold(
+        self,
+        result_entry: Union[
+            RankedPhEvalGeneResult, RankedPhEvalDiseaseResult, RankedPhEvalVariantResult
+        ],
+    ) -> int:
+        """
+        Record the prioritisation rank if it meets the score threshold.
+
+        Args:
+            result_entry (Union[RankedPhEvalGeneResult, RankedPhEvalDiseaseResult, RankedPhEvalVariantResult]):
+                Ranked PhEval result entry
+
+        Returns:
+            int: Recorded prioritisation rank
+        """
+        if float(self.threshold) < float(result_entry.score):
+            return result_entry.rank
+        else:
+            return 0
+
+    def _record_matched_entity(
+        self,
+        standardised_result: Union[
+            RankedPhEvalGeneResult, RankedPhEvalDiseaseResult, RankedPhEvalVariantResult
+        ],
+    ) -> int:
+        """
+        Return the rank result - handling the specification of a threshold.
+        Args:
+            standardised_result (Union[RankedPhEvalGeneResult, RankedPhEvalDiseaseResult, RankedPhEvalVariantResult]):
+                Ranked PhEval disease result entry
+
+        Returns:
+            int: Recorded entity prioritisation rank
+        """
+        if float(self.threshold) == 0.0:
+            return standardised_result.rank
+        else:
+            return (
+                self._assess_with_threshold(standardised_result)
+                if self.score_order != "ascending"
+                else self._assess_with_threshold_ascending_order(
+                    standardised_result,
+                )
+            )

pheval/analyse/benchmark_db_manager.py

@@ -0,0 +1,140 @@
+import ast
+import re
+from typing import List, Type, Union
+
+import duckdb
+from duckdb import DuckDBPyConnection
+
+from pheval.post_processing.post_processing import (
+    RankedPhEvalDiseaseResult,
+    RankedPhEvalGeneResult,
+    RankedPhEvalVariantResult,
+)
+
+
+class BenchmarkDBManager:
+    """
+    Class to connect to database.
+    """
+
+    def __init__(self, benchmark_name: str):
+        """Initialise the BenchmarkDBManager class."""
+        self.conn = self.get_connection(
+            f"{benchmark_name}" if str(benchmark_name).endswith(".db") else f"{benchmark_name}.db"
+        )
+
+    def initialise(self):
+        """Initialise the duckdb connection."""
+        self.add_contains_function()
+
+    @staticmethod
+    def get_connection(db_name: str) -> DuckDBPyConnection:
+        """
+        Get a connection to the database.
+        Returns:
+            DuckDBPyConnection: Connection to the database.
+        """
+        conn = duckdb.connect(db_name)
+        return conn
+
+    def add_column_integer_default(self, table_name: str, column: str, default: int = 0) -> None:
+        """
+        Add a column to an existing table with an integer default value.
+        Args:
+            table_name (str): Name of the table.
+            column (str): Name of the column to add.
+            default (int): Default integer value to add.
+        """
+        try:
+            self.conn.execute(
+                f'ALTER TABLE {table_name} ADD COLUMN "{column}" INTEGER DEFAULT {default}'
+            )
+            self.conn.execute(f'UPDATE {table_name} SET "{column}" = ?', (default,))
+            self.conn.commit()
+        except duckdb.CatalogException:
+            pass
+
+    def drop_table(self, table_name: str) -> None:
+        """
+        Drop a table from the database.
+        Args:
+            table_name: Name of the table to drop from the database
+        """
+        self.conn.execute(f"""DROP TABLE IF EXISTS "{table_name}";""")
+
+    @staticmethod
+    def contains_entity_function(entity: str, known_causative_entity: str) -> bool:
+        """
+        Determines if a known causative entity is present within an entity or list of entities.
+        Args:
+            entity (str): The entity to be checked. It can be a single entity or a string representation of a list.
+            known_causative_entity (str): The entity to search for within the `entity`.
+
+        Returns:
+            bool: `True` if `known_causative_entity` is found in `entity` (or its list representation),
+                `False` otherwise.
+        """
+        list_pattern = re.compile(r"^\[\s*(?:[^\[\],\s]+(?:\s*,\s*[^\[\],\s]+)*)?\s*]$")
+        if list_pattern.match(str(entity)):
+            list_representation = ast.literal_eval(entity)
+            if isinstance(list_representation, list):
+                return known_causative_entity in list_representation
+        return known_causative_entity == entity
+
+    def add_contains_function(self) -> None:
+        """
+        Adds a custom `contains_entity_function` to the DuckDB connection if it does not already exist.
+        """
+        result = self.conn.execute(
+            "SELECT * FROM duckdb_functions() WHERE function_name = ?", ["contains_entity_function"]
+        ).fetchall()
+        if not result:
+            self.conn.create_function("contains_entity_function", self.contains_entity_function)
+
+    def parse_table_into_dataclass(
+        self,
+        table_name: str,
+        dataclass: Union[
+            Type[RankedPhEvalGeneResult],
+            Type[RankedPhEvalVariantResult],
+            Type[RankedPhEvalDiseaseResult],
+        ],
+    ) -> Union[
+        List[RankedPhEvalGeneResult],
+        List[RankedPhEvalVariantResult],
+        List[RankedPhEvalDiseaseResult],
+    ]:
+        """
+        Parses a DuckDB table into a list of dataclass instances.
+        Args:
+            table_name (str): The name of the DuckDB table to be parsed.
+            dataclass (Union[Type[RankedPhEvalGeneResult], Type[RankedPhEvalVariantResult],
+            Type[RankedPhEvalDiseaseResult]]):
+                The dataclass type to which each row in the table should be mapped.
+
+        Returns:
+            List[dataclass]: A list of instances of the provided dataclass, each representing a row from the table.
+        """
+        result = (
+            self.conn.execute(f"SELECT * FROM '{table_name}'").fetchdf().to_dict(orient="records")
+        )
+        return [dataclass(**row) for row in result]
+
+    def check_table_exists(self, table_name: str) -> bool:
+        """
+        Check if a table exists in the connected DuckDB database.
+        Args:
+            table_name (str): The name of the table to check for existence.
+        Returns:
+            bool: Returns `True` if the table exists in the database, `False` otherwise.
+        """
+        result = self.conn.execute(
+            f"SELECT * FROM information_schema.tables WHERE table_name = '{table_name}'"
+        ).fetchall()
+        if result:
+            return True
+        return False
+
+    def close(self):
+        """Close the connection to the database."""
+        self.conn.close()

pheval/analyse/benchmark_generator.py

@@ -1,20 +1,11 @@
-from collections import defaultdict
 from dataclasses import dataclass
 from typing import Callable
 
 from pheval.analyse.benchmarking_data import BenchmarkRunResults
 from pheval.analyse.disease_prioritisation_analysis import benchmark_disease_prioritisation
 from pheval.analyse.gene_prioritisation_analysis import benchmark_gene_prioritisation
-from pheval.analyse.run_data_parser import TrackInputOutputDirectories
+from pheval.analyse.run_data_parser import RunConfig, SinglePlotCustomisation
 from pheval.analyse.variant_prioritisation_analysis import benchmark_variant_prioritisation
-from pheval.constants import (
-    DISEASE_PLOT_FILE_PREFIX,
-    DISEASE_PLOT_Y_LABEL,
-    GENE_PLOT_FILE_PREFIX,
-    GENE_PLOT_Y_LABEL,
-    VARIANT_PLOT_FILE_PREFIX,
-    VARIANT_PLOT_Y_LABEL,
-)
 
 
 @dataclass
@@ -22,20 +13,20 @@ class BenchmarkRunOutputGenerator:
     """Base class for recording data required for generating benchmarking outputs.
 
     Attributes:
-        prioritisation_type_file_prefix (str): Prefix for the prioritisation type output file.
+        plot_customisation (SinglePlotCustomisation): Customisation for plot.
+        prioritisation_type_string (str):  Prioritisation type string.
         y_label (str): Label for the y-axis in benchmarking outputs.
         generate_benchmark_run_results (Callable): Callable to generate benchmark run results.
             Takes parameters: input and results directory, score order, threshold, rank comparison,
             and returns BenchmarkRunResults.
-        stats_comparison_file_suffix (str): Suffix for the rank comparison file.
+        stats_comparison_file (str): Suffix for the rank comparison file.
     """
 
-    prioritisation_type_file_prefix: str
+    plot_customisation: SinglePlotCustomisation
+    prioritisation_type_string: str
     y_label: str
-    generate_benchmark_run_results: Callable[
-        [TrackInputOutputDirectories, str, float, defaultdict], BenchmarkRunResults
-    ]
-    stats_comparison_file_suffix: str
+    generate_benchmark_run_results: Callable[[str, RunConfig, str, float], BenchmarkRunResults]
+    stats_comparison_file: str
 
 
 @dataclass
@@ -48,24 +39,26 @@ class GeneBenchmarkRunOutputGenerator(BenchmarkRunOutputGenerator):
     specifically for gene prioritisation benchmarking.
 
     Attributes:
-        prioritisation_type_file_prefix (str): Prefix for the gene prioritisation type file.
-            Defaults to GENE_PLOT_FILE_PREFIX.
+        plot_customisation (SinglePlotCustomisation): Customisation for plot.
+        prioritisation_type_string (str): Prioritisation type string.
+            Defaults to GENE_PRIORITISATION_TYPE_STR.
         y_label (str): Label for the y-axis in gene prioritisation benchmarking outputs.
             Defaults to GENE_PLOT_Y_LABEL.
         generate_benchmark_run_results (Callable): Callable to generate gene prioritisation
             benchmark run results. Defaults to benchmark_gene_prioritisation.
-            Takes parameters: input and results directory, score order, threshold, rank comparison,
+            Takes parameters: run configuration, score order, threshold, rank comparison,
             and returns BenchmarkRunResults.
-        stats_comparison_file_suffix (str): Suffix for the gene rank comparison file.
-            Defaults to "-gene_summary.tsv".
+        stats_comparison_file (str): Suffix for the gene rank comparison file.
+            Defaults to "-gene_summary".
     """
 
-    prioritisation_type_file_prefix: str = GENE_PLOT_FILE_PREFIX
-    y_label: str = GENE_PLOT_Y_LABEL
-    generate_benchmark_run_results: Callable[
-        [TrackInputOutputDirectories, str, float, defaultdict], BenchmarkRunResults
-    ] = benchmark_gene_prioritisation
-    stats_comparison_file_suffix: str = "-gene_summary.tsv"
+    plot_customisation: SinglePlotCustomisation = None
+    prioritisation_type_string: str = "gene"
+    y_label: str = "Disease-causing genes (%)"
+    generate_benchmark_run_results: Callable[[str, RunConfig, str, float], BenchmarkRunResults] = (
+        benchmark_gene_prioritisation
+    )
+    stats_comparison_file: str = "gene_summary"
 
 
 @dataclass
@@ -78,25 +71,27 @@ class VariantBenchmarkRunOutputGenerator(BenchmarkRunOutputGenerator):
     specifically for variant prioritisation benchmarking.
 
     Attributes:
-        prioritisation_type_file_prefix (str): Prefix for the variant prioritisation type file.
-            Defaults to VARIANT_PLOT_FILE_PREFIX.
+        plot_customisation (SinglePlotCustomisation): Customisation for plot.
+        prioritisation_type_string (str): Prioritisation type string.
+            Defaults to VARIANT_PRIORITISATION_TYPE_STR.
         y_label (str): Label for the y-axis in variant prioritisation benchmarking outputs.
             Defaults to VARIANT_PLOT_Y_LABEL.
         generate_benchmark_run_results (Callable): Callable to generate variant prioritisation
             benchmark run results. Defaults to benchmark_variant_prioritisation.
-            Takes parameters: input and results directory, score order, threshold, rank comparison,
+            Takes parameters: run configuration, score order, threshold, rank comparison,
             and returns BenchmarkRunResults.
-        stats_comparison_file_suffix (str): Suffix for the variant rank comparison file.
-            Defaults to "-variant_summary.tsv".
+        stats_comparison_file (str): Suffix for the variant rank comparison file.
+            Defaults to "-variant_summary".
 
     """
 
-    prioritisation_type_file_prefix: str = VARIANT_PLOT_FILE_PREFIX
-    y_label: str = VARIANT_PLOT_Y_LABEL
-    generate_benchmark_run_results: Callable[
-        [TrackInputOutputDirectories, str, float, defaultdict], BenchmarkRunResults
-    ] = benchmark_variant_prioritisation
-    stats_comparison_file_suffix: str = "-variant_summary.tsv"
+    plot_customisation: SinglePlotCustomisation = None
+    prioritisation_type_string: str = "variant"
+    y_label: str = "Disease-causing variants (%)"
+    generate_benchmark_run_results: Callable[[str, RunConfig, str, float], BenchmarkRunResults] = (
+        benchmark_variant_prioritisation
+    )
+    stats_comparison_file: str = "variant_summary"
 
 
 @dataclass
@@ -109,21 +104,23 @@ class DiseaseBenchmarkRunOutputGenerator(BenchmarkRunOutputGenerator):
     specifically for disease prioritisation benchmarking.
 
     Attributes:
-        prioritisation_type_file_prefix (str): Prefix for the disease prioritisation type file.
-            Defaults to DISEASE_PLOT_FILE_PREFIX.
+        plot_customisation (SinglePlotCustomisation): Customisation for plot.
+        prioritisation_type_string (str): Prioritisation type string.
+            Defaults to DISEASE_PRIORITISATION_TYPE_STR.
         y_label (str): Label for the y-axis in disease prioritisation benchmarking outputs.
             Defaults to DISEASE_PLOT_Y_LABEL.
         generate_benchmark_run_results (Callable): Callable to generate disease prioritisation
             benchmark run results. Defaults to benchmark_disease_prioritisation.
-            Takes parameters: input and results directory, score order, threshold, rank comparison,
+            Takes parameters: run configuration, score order, threshold, rank comparison,
             and returns BenchmarkRunResults.
-        stats_comparison_file_suffix (str): Suffix for the disease rank comparison file.
-            Defaults to "-disease_summary.tsv".
+        stats_comparison_file (str): Suffix for the disease rank comparison file.
+            Defaults to "-disease_summary".
     """
 
-    prioritisation_type_file_prefix: str = DISEASE_PLOT_FILE_PREFIX
-    y_label: str = DISEASE_PLOT_Y_LABEL
-    generate_benchmark_run_results: Callable[
-        [TrackInputOutputDirectories, str, float, defaultdict], BenchmarkRunResults
-    ] = benchmark_disease_prioritisation
-    stats_comparison_file_suffix: str = "-disease_summary.tsv"
+    plot_customisation: SinglePlotCustomisation = None
+    prioritisation_type_string: str = "disease"
+    y_label: str = "Known diseases (%)"
+    generate_benchmark_run_results: Callable[[str, RunConfig, str, float], BenchmarkRunResults] = (
+        benchmark_disease_prioritisation
+    )
+    stats_comparison_file: str = "disease_summary"