Handle empty mutation cohorts without failing

oncodriveclustl/main.py

@@ -282,7 +282,24 @@ def main(input_file,
     logger.info('Validated elements with mutations: {}'.format(elem))
     logger.info('Total substitution mutations: {}'.format(mut))
     if not element_mutations_cutoff:
-        raise excep.UserInputError('No element found with enough mutations to perform analysis')
+        logger.warning('No element found with enough mutations to perform analysis. Writing empty results.')
+        sorted_list_elements = postp.write_element_results(
+            genome=genome,
+            results=({}, {}),
+            directory=output_directory,
+            file=elements_output_file,
+            is_gzip=gzip
+        )
+        postp.write_cluster_results(
+            genome=genome,
+            results=({}, {}),
+            directory=output_directory,
+            file=clusters_output_file,
+            sorter=sorted_list_elements,
+            is_gzip=gzip
+        )
+        logger.info('Finished')
+        return
 
     # Signature
     file_prefix = input_file.split('/')[-1].split('.')[0]

oncodriveclustl/utils/postprocessing.py

@@ -71,6 +71,31 @@ def write_element_results(genome, results, directory, file, is_gzip):
               'P_ANALYTICAL',
               'P_TOPCLUSTER',
               'CGC']
+    column_order = ['SYMBOL',
+                    'ENSID',
+                    'CGC',
+                    'CHROMOSOME',
+                    'STRAND',
+                    'LENGTH',
+                    'TOTAL_MUT',
+                    'CLUSTERED_MUT',
+                    'CLUSTERS',
+                    'SIM_CLUSTERS',
+                    'SCORE',
+                    'P_EMPIRICAL',
+                    'Q_EMPIRICAL',
+                    'P_ANALYTICAL',
+                    'Q_ANALYTICAL',
+                    'P_TOPCLUSTER',
+                    'Q_TOPCLUSTER']
+
+    if not elements_results:
+        df = pd.DataFrame(columns=column_order)
+        if is_gzip is True:
+            df.to_csv(path_or_buf=file, sep='\t', na_rep='', index=False, compression='gzip')
+        else:
+            df.to_csv(path_or_buf=file, sep='\t', na_rep='', index=False)
+        return []
 
     df = pd.DataFrame(columns=header, index=[i for i in range(len(elements_results))])
 
@@ -119,23 +144,7 @@ def write_element_results(genome, results, directory, file, is_gzip):
     df = pd.concat([df_nonempty, df_empty])
 
     # Reorder columns
-    df = df[['SYMBOL',
-             'ENSID',
-             'CGC',
-             'CHROMOSOME',
-             'STRAND',
-             'LENGTH',
-             'TOTAL_MUT',
-             'CLUSTERED_MUT',
-             'CLUSTERS',
-             'SIM_CLUSTERS',
-             'SCORE',
-             'P_EMPIRICAL',
-             'Q_EMPIRICAL',
-             'P_ANALYTICAL',
-             'Q_ANALYTICAL',
-             'P_TOPCLUSTER',
-             'Q_TOPCLUSTER']]
+    df = df[column_order]
 
     # Create a sorted list of elements to order the clusters file
     df.sort_values(by=['Q_ANALYTICAL', 'P_ANALYTICAL', 'SCORE', 'CGC'],