geney 1.2.20__py2.py3-none-any.whl → 1.2.21__py2.py3-none-any.whl

geney/analyzers/visualize_protein_conservation.py

@@ -1,398 +0,0 @@
-
-import pandas as pd
-from collections import namedtuple
-import re
-from geney.oncosplice import oncosplice
-from geney.utils import unload_pickle
-
-def create_figure_story(epistasis, to_file=None):
-    g = epistasis.split(':')[0]
-    out = oncosplice(epistasis, annotate=True)
-    out = out[out.cons_available==1]
-
-    for _, row in out.iterrows():
-        max_length = 0
-        pos = 0
-        for i, k in row.deletions.items():
-            if len(k) > max_length:
-                pos = i
-                max_length = len(k)
-
-        if max_length > 5:
-            del_reg = [pos, pos + max_length]
-        else:
-            del_reg = None
-
-        if row.oncosplice_score == 0:
-            mutation_loc = None
-        else:
-            mutation_loc = pos
-
-        plot_conservation(tid=row.transcript_id,
-                          gene=f'{g}, {row.transcript_id}.{row.isoform}',
-                          mutation_loc=mutation_loc,
-                          target_region=del_reg, mut_name='Epistasis',
-                          domain_annotations=get_annotations(row.transcript_id, 300),
-                          to_file=to_file)
-
-
-import numpy as np
-import matplotlib.pyplot as plt
-from matplotlib.patches import Rectangle
-import seaborn as sns
-
-def plot_conservation(gene_name, tid, gene='', mutation_loc=None, target_region=None, mut_name='Mutation', domain_annotations=[]):
-    """
-    Plots conservation vectors with protein domain visualization and Rate4Site scores.
-
-    Parameters:
-    tid (str): Transcript identifier.
-    gene (str): Gene name.
-    mutation_loc (int): Position of the mutation.
-    target_region (tuple): Start and end positions of the target region.
-    mut_name (str): Name of the mutation.
-    domain_annotations (list): List of tuples for domain annotations (start, end, label).
-    """
-    # Access conservation data
-    _, cons_vec = unload_pickle(gene_name)['tid']['cons_vector']
-
-    if not cons_vec:
-        raise ValueError("The conservation vector is empty.")
-
-    sns.set_theme(style="white")
-    fig, ax = plt.subplots(figsize=(max(15, len(cons_vec)/10), 3))  # Dynamic figure size
-
-    # Plotting the conservation vectors in the main plot
-    plot_conservation_vectors(ax, cons_vec)
-
-    # Setting up primary axis for the main plot
-    setup_primary_axis(ax, gene, len(cons_vec))
-
-    # Create a separate axes for protein domain visualization
-    domain_ax = create_domain_axes(fig, len(cons_vec))
-
-    # Draw protein domains
-    plot_protein_domains(domain_ax, domain_annotations, len(cons_vec))
-
-    # Plotting Rate4Site scores on secondary y-axis
-    plot_rate4site_scores(ax, cons_vec)
-
-    # Plotting mutation location and target region, if provided
-    plot_mutation_and_target_region(ax, mutation_loc, target_region, mut_name)
-
-    plt.show()
-
-def plot_conservation_vectors(ax, cons_vec):
-    """Plots transformed conservation vectors."""
-    temp = transform_conservation_vector(cons_vec, 76)  # Larger window
-    temp /= max(temp)
-    ax.plot(list(range(len(temp))), temp, c='b', label='Estimated Functional Residues')
-
-    temp = transform_conservation_vector(cons_vec, 6)   # Smaller window
-    temp /= max(temp)
-    ax.plot(list(range(len(temp))), temp, c='k', label='Estimated Functional Domains')
-
-def setup_primary_axis(ax, gene, length):
-    """Configures the primary axis of the plot."""
-    ax.set_xlabel(f'AA Position - {gene}', weight='bold')
-    ax.set_xlim(0, length)
-    ax.set_ylim(0, 1)
-    ax.set_ylabel('Relative Importance', weight='bold')
-    ax.tick_params(axis='y')
-    ax.spines['right'].set_visible(False)
-    ax.spines['top'].set_visible(False)
-
-def create_domain_axes(fig, length):
-    """Creates an axis for protein domain visualization."""
-    domain_ax_height = 0.06
-    domain_ax = fig.add_axes([0.125, 0.95, 0.775, domain_ax_height])
-    domain_ax.set_xlim(0, length)
-    domain_ax.set_xticks([])
-    domain_ax.set_yticks([])
-    for spine in domain_ax.spines.values():
-        spine.set_visible(False)
-    return domain_ax
-
-def plot_protein_domains(ax, domain_annotations, length):
-    """Plots protein domain annotations."""
-    ax.add_patch(Rectangle((0, 0), length, 0.9, facecolor='lightgray', edgecolor='none'))
-    for domain in domain_annotations:
-        start, end, label = domain
-        ax.add_patch(Rectangle((start, 0), end - start, 0.9, facecolor='orange', edgecolor='none', alpha=0.5))
-        ax.text((start + end) / 2, 2.1, label, ha='center', va='center', color='black', size=8)
-
-def plot_rate4site_scores(ax, cons_vec):
-    """Plots Rate4Site scores on a secondary y-axis."""
-    ax2 = ax.twinx()
-    c = np.array(cons_vec)
-    c = c + abs(min(c))
-    c = c/max(c)
-    ax2.set_ylim(min(c), max(c)*1.1)
-    ax2.scatter(list(range(len(c))), c, color='green', label='Rate4Site Scores', alpha=0.4)
-    ax2.set_ylabel('Rate4Site Normalized', color='green', weight='bold')
-    ax2.tick_params(axis='y', labelcolor='green')
-    ax2.spines['right'].set_visible(True)
-    ax2.spines['top'].set_visible(False)
-
-def plot_mutation_and_target_region(ax, mutation_loc, target_region, mut_name):
-    """Highlights mutation location and target region, if provided."""
-    if mutation_loc is not None:
-        ax.axvline(x=mutation_loc, ymax=1, color='r', linestyle='--', alpha=0.7)
-        ax.text(mutation_loc, 1.04, mut_name, color='r', weight='bold', ha='center')
-
-    if target_region is not None:
-        ax.add_patch(Rectangle((target_region[0], 0), target_region[1] - target_region[0], 1, alpha=0.25, facecolor='gray'))
-        center_loc = target_region[0] + 0.5 * (target_region[1] - target_region[0])
-        ax.text(center_loc, 1.04, 'Deleted Region', ha='center', va='center', color='gray', weight='bold')
-
-
-# def plot_conservation(tid, gene='', mutation_loc=None, target_region=None, mut_name='Mutation', domain_annotations=[], to_file=None):
-#     _, cons_vec = access_conservation_data(tid)
-#
-#     sns.set_theme(style="white")
-#     fig, ax = plt.subplots(figsize=(15, 3))  # Adjusted figure size for better layout
-#
-#     # Plotting the conservation vectors in the main plot
-#     temp = transform_conservation_vector(cons_vec, 76)
-#     temp /= max(temp)
-#     ax.plot(list(range(len(temp))), temp, c='b', label='Estimated Functional Residues')
-#     temp = transform_conservation_vector(cons_vec, 6)
-#     temp /= max(temp)
-#     ax.plot(list(range(len(temp))), temp, c='k', label='Estimated Functional Domains')
-#
-#     # Setting up primary axis for the main plot
-#     ax.set_xlabel(f'AA Position - {gene}', weight='bold')
-#     ax.set_xlim(0, len(cons_vec))
-#     ax.set_ylim(0, 1)  # Set y-limit to end at 1
-#     ax.set_ylabel('Relative Importance', weight='bold')
-#     ax.tick_params(axis='y')
-#     ax.spines['right'].set_visible(False)
-#     ax.spines['top'].set_visible(False)
-#
-#     # Create a separate axes for protein domain visualization above the main plot
-#     domain_ax_height = 0.06  # Adjust for thinner protein diagram
-#     domain_ax = fig.add_axes([0.125, 0.95, 0.775, domain_ax_height])  # Position higher above the main plot
-#     domain_ax.set_xlim(0, len(cons_vec))
-#     domain_ax.set_xticks([])
-#     domain_ax.set_yticks([])
-#     domain_ax.spines['top'].set_visible(False)
-#     domain_ax.spines['right'].set_visible(False)
-#     domain_ax.spines['left'].set_visible(False)
-#     domain_ax.spines['bottom'].set_visible(False)
-#
-#     # Draw the full-length protein as a base rectangle
-#     domain_ax.add_patch(Rectangle((0, 0), len(cons_vec), 0.9, facecolor='lightgray', edgecolor='none'))
-#
-#     # Overlay domain annotations
-#     for domain in domain_annotations:
-#         start, end, label = domain
-#         domain_ax.add_patch(Rectangle((start, 0), end - start, 0.9, facecolor='orange', edgecolor='none', alpha=0.5))
-#         domain_ax.text((start + end) / 2, 2.1, label, ha='center', va='center', color='black', size=8)
-#
-#     # Plotting Rate4Site scores on secondary y-axis
-#     ax2 = ax.twinx()
-#     c = np.array(cons_vec)
-#     c = c + abs(min(c))
-#     c = c/max(c)
-#     ax2.set_ylim(min(c), max(c)*1.1)
-#     ax2.scatter(list(range(len(c))), c, color='green', label='Rate4Site Scores', alpha=0.4)
-#     ax2.set_ylabel('Rate4Site Normalized', color='green', weight='bold')
-#     ax2.tick_params(axis='y', labelcolor='green')
-#     ax2.spines['right'].set_visible(True)
-#     ax2.spines['top'].set_visible(False)
-#
-#     # Plotting mutation location and target region
-#     if mutation_loc is not None:
-#         ax.axvline(x=mutation_loc, ymax=1,color='r', linestyle='--', alpha=0.7)
-#         ax.text(mutation_loc, 1.04, mut_name, color='r', weight='bold', ha='center')
-#
-#     if target_region is not None:
-#         ax.add_patch(Rectangle((target_region[0], 0), target_region[1] - target_region[0], 1, alpha=0.25, facecolor='gray'))
-#         center_loc = target_region[0] + 0.5 * (target_region[1] - target_region[0])
-#         ax.text(center_loc, 1.04, 'Deleted Region', ha='center', va='center', color='gray', weight='bold')
-#
-#     plt.show()
-#
-
-def merge_overlapping_regions(df):
-    """
-    Merges overlapping regions in a DataFrame.
-
-    Parameters:
-    df (pd.DataFrame): DataFrame with columns 'start', 'end', 'name'
-
-    Returns:
-    List: List of merged regions as namedtuples (start, end, combined_name)
-    """
-    if df.empty:
-        return []
-
-    Region = namedtuple('Region', ['start', 'end', 'combined_name'])
-    df = df.sort_values(by='start')
-    merged_regions = []
-    current_region = None
-
-    for _, row in df.iterrows():
-        start, end, name = row['start'], row['end'], row['name'].replace('_', ' ')
-        if current_region is None:
-            current_region = Region(start, end, [name])
-        elif start <= current_region.end:
-            current_region = Region(current_region.start, max(current_region.end, end), current_region.combined_name + [name])
-        else:
-            merged_regions.append(current_region._replace(combined_name=', '.join(current_region.combined_name)))
-            current_region = Region(start, end, [name])
-
-    if current_region:
-        merged_regions.append(current_region._replace(combined_name=', '.join(current_region.combined_name)))
-
-    # Assuming split_text is a function that splits the text appropriately.
-    merged_regions = [Region(a, b, split_text(c, 35)) for a, b, c in merged_regions]
-    return merged_regions
-
-
-def split_text(text, width):
-    """
-    Splits a text into lines with a maximum specified width.
-
-    Parameters:
-    text (str): The text to be split.
-    width (int): Maximum width of each line.
-
-    Returns:
-    str: The text split into lines of specified width.
-    """
-    lines = re.findall('.{1,' + str(width) + '}', text)
-    return '\n'.join(lines)
-
-
-
-###
-
-
-
-# def plot_conservation(tid, gene='', mutation_loc=None, target_region=None, mut_name='Mutation', domain_annotations=[]):
-#     _, cons_vec = access_conservation_data(tid)
-#
-#     sns.set_theme(style="white")
-#     fig, ax = plt.subplots(figsize=(15, 3))  # Adjusted figure size for better layout
-#
-#     # Plotting the conservation vectors in the main plot
-#     temp = transform_conservation_vector(cons_vec, 76)
-#     temp /= max(temp)
-#     ax.plot(list(range(len(temp))), temp, c='b', label='Estimated Functional Residues')
-#     temp = transform_conservation_vector(cons_vec, 7)
-#     temp /= max(temp)
-#     ax.plot(list(range(len(temp))), temp, c='k', label='Estimated Functional Domains')
-#
-#     # Setting up primary axis for the main plot
-#     ax.set_xlabel(f'AA Position - {gene}', weight='bold')
-#     ax.set_xlim(0, len(cons_vec))
-#     ax.set_ylim(0, 1)  # Set y-limit to end at 1
-#     ax.set_ylabel('Relative Importance', weight='bold')
-#     ax.tick_params(axis='y')
-#     ax.spines['right'].set_visible(False)
-#     ax.spines['top'].set_visible(False)
-#
-#     # Create a separate axes for protein domain visualization above the main plot
-#     domain_ax_height = 0.06  # Adjust for thinner protein diagram
-#     domain_ax = fig.add_axes([0.125, 0.95, 0.775, domain_ax_height])  # Position higher above the main plot
-#     domain_ax.set_xlim(0, len(cons_vec))
-#     domain_ax.set_xticks([])
-#     domain_ax.set_yticks([])
-#     domain_ax.spines['top'].set_visible(False)
-#     domain_ax.spines['right'].set_visible(False)
-#     domain_ax.spines['left'].set_visible(False)
-#     domain_ax.spines['bottom'].set_visible(False)
-#
-#     # Draw the full-length protein as a base rectangle
-#     domain_ax.add_patch(Rectangle((0, 0), len(cons_vec), 0.9, facecolor='lightgray', edgecolor='none'))
-#
-#     # Overlay domain annotations
-#     for domain in domain_annotations:
-#         start, end, label = domain
-#         domain_ax.add_patch(Rectangle((start, 0), end - start, 0.9, facecolor='orange', edgecolor='none', alpha=0.5))
-#         domain_ax.text((start + end) / 2, 2.1, label, ha='center', va='center', color='black', size=8)
-#
-#     # Plotting Rate4Site scores on secondary y-axis
-#     ax2 = ax.twinx()
-#     c = np.array(cons_vec)
-#     c = c + abs(min(c))
-#     c = c/max(c)
-#     ax2.set_ylim(min(c), max(c)*1.1)
-#     ax2.scatter(list(range(len(c))), c, color='green', label='Rate4Site Scores', alpha=0.4)
-#     ax2.set_ylabel('Rate4Site Normalized', color='green', weight='bold')
-#     ax2.tick_params(axis='y', labelcolor='green')
-#     ax2.spines['right'].set_visible(True)
-#     ax2.spines['top'].set_visible(False)
-#
-#     # Plotting mutation location and target region
-#     if mutation_loc is not None:
-#         ax.axvline(x=mutation_loc, ymax=1,color='r', linestyle='--', alpha=0.7)
-#         ax.text(mutation_loc, 1.01, mut_name, color='r', weight='bold', ha='center')
-#
-#     if target_region is not None:
-#         ax.add_patch(Rectangle((target_region[0], 0.85), target_region[1] - target_region[0], 0.05, alpha=0.3, facecolor='blue'))
-#         center_loc = target_region[0] + 0.5 * (target_region[1] - target_region[0])
-#         ax.text(center_loc, 0.875, 'Target Region', ha='center', va='center', color='blue', weight='bold')
-#
-#     plt.show()
-#
-#
-# def merge_overlapping_regions(df):
-#     # Sort the DataFrame by the 'start' column
-#     df = df.sort_values(by='start')
-#
-#     merged_regions = []  # List to store merged regions as tuples (start, end, combined_name)
-#
-#     current_start = None
-#     current_end = None
-#     combined_names = []  # List to store names of overlapping regions
-#
-#     for index, row in df.iterrows():
-#         start = row['start']
-#         end = row['end']
-#         name = row['name'].replace('_', ' ')
-#
-#         if current_start is None:
-#             # Initialize the current region
-#             current_start = start
-#             current_end = end
-#             combined_names.append(name)
-#         else:
-#             if start <= current_end:
-#                 # Regions overlap, update the current region and add the name to combined_names
-#                 current_end = max(current_end, end)
-#                 combined_names.append(name)
-#             else:
-#                 # Regions don't overlap, add the current region to the result with combined names
-#                 combined_name = ', '.join(combined_names)
-#                 merged_regions.append((current_start, current_end, combined_name))
-#                 # Start a new current region with the current row
-#                 current_start = start
-#                 current_end = end
-#                 combined_names = [name]
-#
-#     # Add the last current region to the result
-#     if current_start is not None:
-#         combined_name = ', '.join(combined_names)
-#         merged_regions.append((current_start, current_end, combined_name))
-#
-#     merged_regions = [(a, b, split_text(c, 35)) for a, b, c in merged_regions]
-#     return merged_regions
-#
-# def split_text(text, width):
-#     lines = []
-#     while text:
-#         # Find the index to split at or take the whole text if it's shorter than the width
-#         split_index = min(len(text), width)
-#         # Append the substring up to the split index to the lines list
-#         lines.append(text[:split_index])
-#         # Remove the processed substring from the original text
-#         text = text[split_index:]
-#     return '\n'.join(lines)
-#
-# def get_annotations(target_gene, w=500):
-#     temp = PROTEIN_ANNOTATIONS[(PROTEIN_ANNOTATIONS['Transcript stable ID'] == PROTEIN_ANNOTATIONS[target_gene]) & (PROTEIN_ANNOTATIONS.length < w)].drop_duplicates(subset=['Interpro Short Description'], keep='first')
-#     return merge_overlapping_regions(temp)
-#
-#

geney/benchmark_clinvar.py

@@ -1,158 +0,0 @@
-import pandas as pd
-from sklearn.metrics import roc_curve, precision_recall_curve
-import matplotlib.pyplot as plt
-from datetime import datetime
-from pathlib import Path
-import subprocess
-
-from geney import config_setup
-from geney.utils import download_and_gunzip
-from geney.oncosplice import oncosplice_reduced
-
-def download_and_parse_clinvar():
-    url = 'https://ftp.ncbi.nlm.nih.gov/pub/clinvar/vcf_GRCh38/clinvar.vcf.gz'
-    local_file = download_and_gunzip(url, target_path)
-    return local_file
-
-
-def aggregate_clinvar_results(benchmark_path, aggregate_mode=False, benchmark_feature=None, local_clinvar_df='/tamir2/nicolaslynn/data/ClinVar/clinvar_compact.csv'):
-    data = pd.concat([pd.read_csv(file) for file in Path(benchmark_path).glob('*.csv')])
-    if not aggregate_mode:
-        data = data[(data.cons_available) & (data.primary_transcript)]
-
-    data = oncosplice_reduced(data)
-    data = data.loc[:, ~data.columns.duplicated()]
-    data = pd.merge(data, pd.read_csv(local_clinvar_df), on='mut_id')
-    data['clinsig_val'] = data.apply(lambda row: {'Benign': 0, 'Pathogenic': 1}[row.clinsig], axis=1)
-    for c in data.columns:
-        try:
-            if data[c].min() < 0:
-                data[f'{c}_abs'] = abs(data[c])
-        except TypeError:
-            pass
-
-    print(data.corr(numeric_only=True))
-    print(data.corrwith(data['clinsig_val'], method='spearman'))
-    print(data.corrwith(data['clinsig_val'], method='pearson'))
-    return data
-
-
-def plot_performance(true_values, predictions):
-    clinsig_map = {'Benign': 0, 'Pathogenic': 1}
-    true_values = [clinsig_map[t] for t in true_values]
-    predictions = scale_predictions(predictions)
-
-    fpr, tpr, thresholds_roc = roc_curve(true_values, predictions)
-
-    # Calculate Precision-Recall curve
-    precision, recall, thresholds_pr = precision_recall_curve(true_values, predictions)
-
-    # Plotting ROC curve
-    plt.figure(figsize=(20, 5))
-
-    plt.subplot(1, 4, 1)
-    plt.plot(fpr, tpr)
-    plt.title('ROC Curve')
-    plt.xlabel('False Positive Rate')
-    plt.ylabel('True Positive Rate')
-
-    # Plotting Precision-Recall curve
-    plt.subplot(1, 4, 2)
-    plt.plot(recall, precision)
-    plt.title('Precision-Recall Curve')
-    plt.xlabel('Recall')
-    plt.ylabel('Precision')
-
-    # Plotting Precision vs. Thresholds
-    plt.subplot(1, 4, 3)
-    plt.plot(thresholds_pr, precision[:-1])  # Precision and thresholds have off-by-one lengths
-    plt.title('Precision vs. Threshold')
-    plt.xlabel('Threshold')
-    plt.ylabel('Precision')
-
-    # Plotting Sample Percentage Captured vs. Thresholds
-    plt.subplot(1, 4, 4)
-    # Assuming 'tpr' or another appropriate metric represents the cumulative percentage
-    plt.plot(thresholds_roc, tpr)  # Update 'tpr' with the correct metric if necessary
-    plt.title('Cumulative Percentage vs. Threshold')
-    plt.xlabel('Threshold')
-    plt.ylabel('Cumulative Percentage of Population')
-
-    plt.tight_layout()
-    plt.show()
-
-
-
-class ClinVarBenchmark:
-    def __init__(self, df):
-        assert 'clinsig' in df.columns, 'No clinsig column found in dataframe.'
-        self.df = df
-
-
-    def scale_predictions(self, p):
-        max_val = max(p)
-        min_val = min(p)
-        return (p - min_val) / (max_val - min_val)
-
-    def plot_performance(self, true_values, predictions):
-        clinsig_map = {'Benign': 0, 'Pathogenic': 1}
-        predictions = [clinsig_map[t] for t in true_values]
-        predictions = self.scale_predictions(predictions)
-
-        fpr, tpr, thresholds_roc = roc_curve(true_values, predictions)
-
-        # Calculate Precision-Recall curve
-        precision, recall, thresholds_pr = precision_recall_curve(true_values, predictions)
-
-        # Plotting ROC curve
-        plt.figure(figsize=(20, 5))
-
-        plt.subplot(1, 4, 1)
-        plt.plot(fpr, tpr)
-        plt.title('ROC Curve')
-        plt.xlabel('False Positive Rate')
-        plt.ylabel('True Positive Rate')
-
-        # Plotting Precision-Recall curve
-        plt.subplot(1, 4, 2)
-        plt.plot(recall, precision)
-        plt.title('Precision-Recall Curve')
-        plt.xlabel('Recall')
-        plt.ylabel('Precision')
-
-        # Plotting Precision vs. Thresholds
-        plt.subplot(1, 4, 3)
-        plt.plot(thresholds_pr, precision[:-1])  # Precision and thresholds have off-by-one lengths
-        plt.title('Precision vs. Threshold')
-        plt.xlabel('Threshold')
-        plt.ylabel('Precision')
-
-        # Plotting Sample Percentage Captured vs. Thresholds
-        plt.subplot(1, 4, 4)
-        # Assuming 'tpr' or another appropriate metric represents the cumulative percentage
-        plt.plot(thresholds_roc, tpr)  # Update 'tpr' with the correct metric if necessary
-        plt.title('Cumulative Percentage vs. Threshold')
-        plt.xlabel('Threshold')
-        plt.ylabel('Cumulative Percentage of Population')
-
-        plt.tight_layout()
-        plt.show()
-        return None
-
-    def report(self, feature):
-        pass
-
-    def find_ppv_threshold(self, feature, ppv_threshold=0.95):
-        pass
-
-
-
-if __name__ ==  '__main__':
-    now = datetime.now()
-    benchmark_path = config_setup['ONCOSPLICE'] / f'clinvar_benchmark_{now.strftime("%m_%d_%Y")}'
-    print(f"Saving benchmark results to {benchmark_path}")
-    benchmark_path.mkdir(parents=True, exist_ok=True)
-    subprocess.run(['python', '-m', 'geney.pipelines.power_utils', '-i',
-                    '/tamir2/nicolaslynn/data/ClinVar/clinvar_oncosplice_input.txt', '-r', str(benchmark_path),
-                    '-n', '10', '-m', '5GB'])
-

geney/compare_sets.py

@@ -1,91 +0,0 @@
-import pandas as pd
-import numpy as np
-from sklearn.metrics import precision_score, recall_score, accuracy_score
-from sklearn.metrics import roc_auc_score, roc_curve
-import matplotlib.pyplot as plt
-
-def plot_auc_curve(y_true, y_pred_proba):
-    """
-    Plots the AUC curve.
-
-    Args:
-        y_true (array-like): True labels (0 or 1).
-        y_pred_proba (array-like): Predicted probabilities for positive class.
-
-    Returns:
-        None
-    """
-    fpr, tpr, _ = roc_curve(y_true, y_pred_proba)
-    auc_value = roc_auc_score(y_true, y_pred_proba)
-
-    plt.figure(figsize=(8, 6))
-    plt.plot(fpr, tpr, label=f"AUC = {auc_value:.2f}")
-    plt.plot([0, 1], [0, 1], 'k--')
-    plt.xlabel("False Positive Rate")
-    plt.ylabel("True Positive Rate")
-    plt.title("Receiver Operating Characteristic (ROC) Curve")
-    plt.legend()
-    plt.show()
-    return auc_value
-
-
-def optimal_ppv(dataframe, feature_name, plot=False):
-    """
-    Calculates the optimal positive predictive value (PPV) for a given feature.
-
-    Args:
-        dataframe (pd.DataFrame): Input dataframe.
-        feature_name (str): Name of the feature column.
-
-    Returns:
-        float: Optimal PPV.
-    """
-    # Assuming 'target' is the binary target column (0 or 1)
-    threshold_values = pd.qcut(dataframe[feature_name], 100, duplicates='drop')
-    ppv_values = []
-
-    for threshold in threshold_values:
-        predictions = (dataframe[feature_name] >= threshold).astype(int)
-        ppv = precision_score(dataframe['target'], predictions)
-        ppv_values.append(ppv)
-
-    optimal_threshold = threshold_values[np.argmax(ppv_values)]
-    optimal_ppv = max(ppv_values)
-    if plot:
-        plt.figure(figsize=(8, 6))
-        plt.scatter(threshold_values, ppv_values)
-        plt.xlabel("Threshold")
-        plt.ylabel("Positive Predictive Value (PPV)")
-        plt.title("Optimal Positive Predictive Value (PPV)")
-        plt.show()
-
-    return optimal_ppv, optimal_threshold
-
-
-def measure_prediction_quality(prediction_vector, quality_vector):
-    """
-    Measure the quality of the predictions using the quality_vector as the characteristic to check.
-    """
-    pass
-
-
-
-def create_ppv_vector(prediction_vector, true_value_vector):
-    """
-    Create a vector of positive predictive values (PPV) for the prediction_vector using the true_value_vector as the true values.
-    """
-    df = pd.DataFrame({'prediction': prediction_vector, 'true_value': true_value_vector})
-    df.sort_values('prediction', ascending=True, inplace=True)
-    df['bin'] = pd.qcut(df['prediction'], 100, labels=False, duplicates=True, retbins=True)
-    for bin in df.bin.unique():
-        temp_df = df[df.bin >= bin].
-
-
-def group_retention(predictions, predictor):
-    # first i need to get the ratio of values that are retained at particular values
-    predictions.sort_values(predictor, inplace=True)
-    _, thresholds = pd.qcut(predictions[predictor], 100, duplicates='drop')
-    tracker = []
-    for th in thresholds:
-
-