fix bug in 16S validtion w premade indices

argparse_dicts.py

@@ -363,7 +363,7 @@ def get_validate_parser():
     for arg in (fast5_in, out_dir, target_16S, nb_folds, cores,
                 parameter_file, hdf_path):
         parser.add_argument(arg[0], **arg[1])
-    parser.add_argument('--nn-dir', required=True, type=lambda x: check_input_path(x),
+    parser.add_argument('--nn-dir', required=False, type=lambda x: check_input_path(x), default=f'{__location__}/16S_db/',
                         help='Directory containing pre-made nns for single k-mers, to be used directly')
     parser.add_argument('--ground-truth-16s', required=True, type=str, help='csv denoting which species reads belong to')
     parser.add_argument('--index-files', type=lambda x: check_input_path(x),

validation/validate_16S.py

@@ -91,9 +91,11 @@ test_read_dir = parse_output_path(f'{args.out_dir}test_reads/', clean=True)
 if args.index_files:
     index_fn_list = sorted(os.listdir(args.index_files))
     for fi, fn in enumerate(index_fn_list):
-        idx_df = pd.read_csv(f'{args.index_files}{fn}',index_col=0)
+        idx_df = pd.read_csv(f'{args.index_files}{fn}', index_col=0)
+        idx_df.loc[:, 'fn'] = args.fast5_in + idx_df.index
         fast5_df.loc[:, f'fold_{fi}'] = idx_df.loc[fast5_df.index, 'fold']
-        copy(f'{args.index_files}{fn}', f'{read_index_dir}index_fold{fi}.csv')
+        idx_df.to_csv(f'{read_index_dir}index_fold{fi}.csv')
+        # copy(f'{args.index_files}{fn}', f'{read_index_dir}index_fold{fi}.csv')
         for sp in species_list:
             _ = parse_output_path(f'{test_read_dir}{sp}/fold_{fi}')
 else: