spacr 0.3.50__py3-none-any.whl → 0.3.55__py3-none-any.whl

spacr/ml.py

@@ -155,10 +155,6 @@ def process_model_coefficients(model, regression_type, X, y, nc, pc, controls):
     coef_df['condition'] = coef_df.apply(lambda row: 'nc' if nc in row['feature'] else 'pc' if pc in row['feature'] else ('control' if row['grna'] in controls else 'other'),axis=1)
     return coef_df[~coef_df['feature'].str.contains('row|column')]
 
-
-
-
-
 def check_distribution(y):
     """Check the type of distribution to recommend a model."""
     if np.all((y == 0) | (y == 1)):
@@ -288,7 +284,7 @@ def check_and_clean_data(df, dependent_variable):
     df = handle_missing_values(df, ['fraction', dependent_variable])
     
     # Step 2: Ensure grna, gene, plate, row, column, and prc are categorical types
-    df = ensure_valid_types(df, ['grna', 'gene', 'plate', 'row', 'column', 'prc'])
+    df = ensure_valid_types(df, ['grna', 'gene', 'plate', 'row_name', 'column', 'prc'])
     
     # Step 3: Check for multicollinearity in fraction and the dependent variable
     df_cleaned = check_collinearity(df, ['fraction', dependent_variable])
@@ -298,7 +294,7 @@ def check_and_clean_data(df, dependent_variable):
     df_cleaned['grna'] = df['grna']
     df_cleaned['prc'] = df['prc']
     df_cleaned['plate'] = df['plate']
-    df_cleaned['row'] = df['row']
+    df_cleaned['row_name'] = df['row_name']
     df_cleaned['column'] = df['column']
 
     # Create a new column 'gene_fraction' that sums the fractions by gene within the same well
@@ -337,7 +333,7 @@ def minimum_cell_simulation(settings, num_repeats=10, sample_size=100, tolerance
         df = pd.read_csv(score_data)
         df = correct_metadata_column_names(df)
         df['plate'] = f'plate{i + 1}'
-        df['prc'] = df['plate'] + '_' + df['row'].astype(str) + '_' + df['column'].astype(str)
+        df['prc'] = df['plate'] + '_' + df['row_name'].astype(str) + '_' + df['column'].astype(str)
         dfs.append(df)
 
     df = pd.concat(dfs, axis=0)
@@ -706,18 +702,16 @@ def perform_regression(settings):
     def _perform_regression_read_data(settings):
 
         if isinstance(settings['score_data'], list) and isinstance(settings['count_data'], list):
-            settings['plate'] = None
             if len(settings['score_data']) == 1:
-                settings['score_data'] = settings['score_data'][0]
-            if len(settings['count_data']) == 1:
-                settings['count_data'] = settings['count_data'][0]
+                count_data_df = pd.read_csv(settings['count_data'][0])
+                score_data_df = pd.read_csv(settings['score_data'][0])
             else:
                 count_data_df = pd.DataFrame()
                 for i, count_data in enumerate(settings['count_data']):
                     df = pd.read_csv(count_data)
                     df['plate_name'] = f'plate{i+1}'
                     if 'column' in df.columns:
-                        df['col'] = df['column']
+                        df['column_name'] = df['column']
                     count_data_df = pd.concat([count_data_df, df])
                     print('Count data:', len(count_data_df))
 
@@ -726,7 +720,7 @@ def perform_regression(settings):
                     df = pd.read_csv(score_data)
                     df['plate_name'] = f'plate{i+1}'
                     if 'column' in df.columns:
-                        df['col'] = df['column']
+                        df['column_name'] = df['column']
                     score_data_df = pd.concat([score_data_df, df])
                     print('Score data:', len(score_data_df))
         else:
@@ -806,9 +800,23 @@ def perform_regression(settings):
             return df, n_gene
         else:
             return df
-        
+
     settings = get_perform_regression_default_settings(settings)
     count_data_df, score_data_df = _perform_regression_read_data(settings)
+    
+    if "row_name" in count_data_df.columns:
+        num_parts = len(count_data_df['row_name'].iloc[0].split('_'))
+        if num_parts == 2:
+            split = count_data_df['row_name'].str.split('_', expand=True)
+            count_data_df['row_name'] = split[1]
+    
+    if "prc" in score_data_df.columns:
+        num_parts = len(score_data_df['prc'].iloc[0].split('_'))
+        if num_parts == 3:
+            split = score_data_df['prc'].str.split('_', expand=True)
+            score_data_df['plate'] = settings['plate']
+            score_data_df['prc'] = score_data_df['plate'] + '_' + split[1] + '_' + split[2]
+        
     results_path, results_path_gene, results_path_grna, hits_path, res_folder, csv_path = _perform_regression_set_paths(settings)
     save_settings(settings, name='regression', show=True)
 
@@ -849,7 +857,7 @@ def perform_regression(settings):
     merged_df.to_csv(data_path, index=False)
     print(f"Saved regression data to {data_path}")
 
-    merged_df[['plate', 'row', 'column']] = merged_df['prc'].str.split('_', expand=True)
+    merged_df[['plate', 'row_name', 'column']] = merged_df['prc'].str.split('_', expand=True)
     
     _ = plot_plates(merged_df, variable=orig_dv, grouping='mean', min_max='allq', cmap='viridis', min_count=None, dst=res_folder)                
 
@@ -857,6 +865,7 @@ def perform_regression(settings):
     
     coef_df['grna'] = coef_df['feature'].apply(lambda x: re.search(r'grna\[(.*?)\]', x).group(1) if 'grna' in x else None)
     coef_df['gene'] = coef_df['feature'].apply(lambda x: re.search(r'gene\[(.*?)\]', x).group(1) if 'gene' in x else None)
+    
     coef_df = coef_df.merge(n_grna, how='left', on='grna')
     coef_df = coef_df.merge(n_gene, how='left', on='gene')
 
@@ -903,7 +912,6 @@ def perform_regression(settings):
         save_summary_to_file(model, file_path=f'{res_folder}/mode_summary.csv')
     
     significant.to_csv(hits_path, index=False)
-
     significant_grna_filtered = significant[significant['n_grna'] > settings['min_n']]
     significant_gene_filtered = significant[significant['n_gene'] > settings['min_n']]
     significant_filtered = pd.concat([significant_grna_filtered, significant_gene_filtered])
@@ -928,8 +936,6 @@ def perform_regression(settings):
         base_dir = os.path.dirname(os.path.abspath(__file__))
         metadata_path = os.path.join(base_dir, 'resources', 'data', 'lopit.csv')
         
-        display(data_path)
-
         if settings['volcano'] == 'all':
             print('all')
             gene_list = custom_volcano_plot(data_path, metadata_path, metadata_column='tagm_location', point_size=600, figsize=20, threshold=reg_threshold, save_path=volcano_path, x_lim=settings['x_lim'],y_lims=settings['y_lims'])
@@ -982,14 +988,14 @@ def process_reads(csv_path, fraction_threshold, plate, filter_column=None, filte
         csv_df = csv_df.rename(columns={'plate_name': 'plate'})
     if 'column_name' in csv_df.columns:
         csv_df = csv_df.rename(columns={'column_name': 'column'})
-    if 'col' in csv_df.columns:
-        csv_df = csv_df.rename(columns={'col': 'column'})
+    if 'column_name' in csv_df.columns:
+        csv_df = csv_df.rename(columns={'column_name': 'column'})
     if 'row_name' in csv_df.columns:
-        csv_df = csv_df.rename(columns={'row_name': 'row'})
+        csv_df = csv_df.rename(columns={'row_name': 'row_name'})
     if 'grna_name' in csv_df.columns:
         csv_df = csv_df.rename(columns={'grna_name': 'grna'})
     if 'plate_row' in csv_df.columns:
-        csv_df[['plate', 'row']] = csv_df['plate_row'].str.split('_', expand=True)
+        csv_df[['plate', 'row_name']] = csv_df['plate_row'].str.split('_', expand=True)
 
     if not 'plate' in csv_df.columns:
         if not plate is None:
@@ -1009,11 +1015,11 @@ def process_reads(csv_path, fraction_threshold, plate, filter_column=None, filte
                 csv_df = csv_df[csv_df[filter_col] != value]
     
     # Ensure the necessary columns are present
-    if not all(col in csv_df.columns for col in ['row','column','grna','count']):
-        raise ValueError("The CSV file must contain 'grna', 'count', 'row', and 'column' columns.")
+    if not all(col in csv_df.columns for col in ['row_name','column','grna','count']):
+        raise ValueError("The CSV file must contain 'grna', 'count', 'row_name', and 'column' columns.")
 
     # Create the prc column
-    csv_df['prc'] = csv_df['plate'] + '_' + csv_df['row'] + '_' + csv_df['column']
+    csv_df['prc'] = csv_df['plate'] + '_' + csv_df['row_name'] + '_' + csv_df['column']
 
     # Group by prc and calculate the sum of counts
     grouped_df = csv_df.groupby('prc')['count'].sum().reset_index()
@@ -1075,7 +1081,7 @@ def clean_controls(df,values, column):
     return df
 
 def process_scores(df, dependent_variable, plate, min_cell_count=25, agg_type='mean', transform=None, regression_type='ols'):
-
+    
     if 'plate_name' in df.columns:
         df.drop(columns=['plate'], inplace=True)
         df = df.rename(columns={'plate_name': 'plate'})
@@ -1083,11 +1089,14 @@ def process_scores(df, dependent_variable, plate, min_cell_count=25, agg_type='m
     if plate is not None:
         df['plate'] = plate
 
-    if 'col' not in df.columns:
-        df['col'] = df['column']
+    if 'column_name' not in df.columns:
+        df['column_name'] = df['column']
 
-    df['prc'] = df['plate'].astype(str) + '_' + df['row'].astype(str) + '_' + df['col'].astype(str)
+    df['prc'] = df['plate'].astype(str) + '_' + df['row_name'].astype(str) + '_' + df['column_name'].astype(str)
 
+    display(df)
+    
+    
     df = df[['prc', dependent_variable]]
 
     # Group by prc and calculate the mean and count of the dependent_variable
@@ -1257,7 +1266,7 @@ def generate_ml_scores(settings):
 
     return [output, plate_heatmap]
 
-def ml_analysis(df, channel_of_interest=3, location_column='col', positive_control='c2', negative_control='c1', exclude=None, n_repeats=10, top_features=30, n_estimators=100, test_size=0.2, model_type='xgboost', n_jobs=-1, remove_low_variance_features=True, remove_highly_correlated_features=True, verbose=False):
+def ml_analysis(df, channel_of_interest=3, location_column='column_name', positive_control='c2', negative_control='c1', exclude=None, n_repeats=10, top_features=30, n_estimators=100, test_size=0.2, model_type='xgboost', n_jobs=-1, remove_low_variance_features=True, remove_highly_correlated_features=True, verbose=False):
     
     """
     Calculates permutation importance for numerical features in the dataframe,
@@ -1403,8 +1412,8 @@ def ml_analysis(df, channel_of_interest=3, location_column='col', positive_contr
     df = _calculate_similarity(df, features, location_column, positive_control, negative_control)
 
     df['prcfo'] = df.index.astype(str)
-    df[['plate', 'row', 'col', 'field', 'object']] = df['prcfo'].str.split('_', expand=True)
-    df['prc'] = df['plate'] + '_' + df['row'] + '_' + df['col']
+    df[['plate', 'row_name', 'column_name', 'field', 'object']] = df['prcfo'].str.split('_', expand=True)
+    df['prc'] = df['plate'] + '_' + df['row_name'] + '_' + df['column_name']
     
     return [df, permutation_df, feature_importance_df, model, X_train, X_test, y_train, y_test, metrics_df], [permutation_fig, feature_importance_fig]