BD facs diva support (#46)

BD facs diva devices only have channel_name_s set for some but not all channels. If not present for all channels fcsparser fallbacks to the channel_names_n.

This pull request includes two main adaptations:

- Using channel_names_n as a fallback for channels without a value in channel_names_s
- Puts both $PnN and $PnS as columns into the metadata dataframe

All adaptations have corresponding unit tests

.gitignore

@@ -2,6 +2,9 @@
 **/*.pyc
 .cache
 /venv
-/build
 /.vscode
-/dist
+/build
+/dist
+versioneer.py
+/fcsparser.egg-info
+setup.cfg

fcsparser/api.py

@@ -375,22 +375,24 @@ def read_text(self, file_handle):
         else:
             self.channel_numbers = range(1, pars + 1)  # Channel numbers start from 1
 
+
         # Extract parameter names
-        try:
-            names_n = tuple([text["$P{0}N".format(i)] for i in self.channel_numbers])
-        except KeyError:
-            names_n = []
-
-        try:
-            names_s = tuple([text["$P{0}S".format(i)] for i in self.channel_numbers])
-        except KeyError:
-            names_s = []
-
-        self.channel_names_s = names_s
-        self.channel_names_n = names_n
-
+        channel_names_n = []
+        channel_names_s = []
+        for channel_number in self.channel_numbers:
+            n_key = f'$P{channel_number}N'
+            s_key = f'$P{channel_number}S'
+            name_n = text.get(n_key, "")
+            name_s = text.get(s_key, "")
+
+            channel_names_n.append(name_n)
+            channel_names_s.append(name_s if name_s != "" else name_n)
+
+        self.channel_names_n = tuple(channel_names_n)
+        self.channel_names_s = tuple(channel_names_s)
+
         # Convert some of the fields into integer values
-        keys_encoding_bits = ["$P{0}B".format(i) for i in self.channel_numbers]
+        keys_encoding_bits = [f"$P{channel_number}B" for channel_number in self.channel_numbers]
 
         add_keys_to_convert_to_int = ["$NEXTDATA", "$PAR", "$TOT"]
 
@@ -642,9 +644,12 @@ def reformat_meta(self):
         if "$PnE" in column_names:
             df["$PnE"] = df["$PnE"].apply(lambda x: x.split(","))
 
-        df.index.name = "Channel Number"
-        meta["_channels_"] = df
-        meta["_channel_names_"] = self.get_channel_names()
+        if self._channel_naming not in df.columns.names:
+            df[self._channel_naming] = self.get_channel_names()
+
+        df.index.name = 'Channel Number'
+        meta['_channels_'] = df
+        meta['_channel_names_'] = self.get_channel_names()
 
     @property
     def dataframe(self):

fcsparser/tests/test_fcs_reader.py

@@ -14,25 +14,24 @@
 BASE_PATH = os.path.join(HERE, 'data', 'FlowCytometers')
 
 # Used for checking data segments in fcs files generated by different machines.
-FILE_IDENTIFIER_TO_PATH = {'mq fcs 2.0': os.path.join(BASE_PATH, 'MiltenyiBiotec', 'FCS2.0',
-                                                      'EY_2013-07-19_PBS_FCS_2.0_Custom_Without_Add_Well_A1.001.fcs'),
-                           'mq fcs 3.0': os.path.join(BASE_PATH, 'MiltenyiBiotec', 'FCS3.0',
-                                                      'FCS3.0_Custom_Compatible.fcs'),
-                           'mq fcs 3.1': os.path.join(BASE_PATH, 'MiltenyiBiotec', 'FCS3.1',
-                                                      'EY_2013-07-19_PBS_FCS_3.1_Well_A1.001.fcs'),
-                           'LSR II fcs 3.0': os.path.join(BASE_PATH, 'HTS_BD_LSR-II',
-                                                          'HTS_BD_LSR_II_Mixed_Specimen_001_D6_D06.fcs'),
-                           'Fortessa fcs 3.0': os.path.join(BASE_PATH, 'Fortessa',
-                                                            'FCS_3.0_Fortessa_PBS_Specimen_001_A1_A01.fcs'),
-                           'large fake fcs': os.path.join(BASE_PATH, 'fake_large_fcs',
-                                                          'fake_large_fcs.fcs'),
-                           'cyflow cube 8': os.path.join(BASE_PATH, 'cyflow_cube_8',
-                                                         'cyflow_cube_8.fcs'),
-                           'fake bitmask error': os.path.join(BASE_PATH, 'fake_bitmask_error',
-                                                              'fcs1_cleaned.lmd'),
-                           'Cytek xP5': os.path.join(BASE_PATH, 'Cytek_xP5', 'Cytek_xP5.fcs'),
-                           'guava muse': os.path.join(BASE_PATH, 'GuavaMuse',
-                                                      'Guava Muse.fcs'), }
+FILE_IDENTIFIER_TO_PATH = {
+    'mq fcs 2.0': os.path.join(BASE_PATH, 'MiltenyiBiotec', 'FCS2.0',
+                               'EY_2013-07-19_PBS_FCS_2.0_Custom_Without_Add_Well_A1.001.fcs'),
+    'mq fcs 3.0': os.path.join(BASE_PATH, 'MiltenyiBiotec', 'FCS3.0',
+                               'FCS3.0_Custom_Compatible.fcs'),
+    'mq fcs 3.1': os.path.join(BASE_PATH, 'MiltenyiBiotec', 'FCS3.1',
+                               'EY_2013-07-19_PBS_FCS_3.1_Well_A1.001.fcs'),
+    'LSR II fcs 3.0': os.path.join(BASE_PATH, 'HTS_BD_LSR-II',
+                                   'HTS_BD_LSR_II_Mixed_Specimen_001_D6_D06.fcs'),
+    'Fortessa fcs 3.0': os.path.join(BASE_PATH, 'Fortessa',
+                                     'FCS_3.0_Fortessa_PBS_Specimen_001_A1_A01.fcs'),
+    'large fake fcs': os.path.join(BASE_PATH, 'fake_large_fcs', 'fake_large_fcs.fcs'),
+    'cyflow cube 8': os.path.join(BASE_PATH, 'cyflow_cube_8', 'cyflow_cube_8.fcs'),
+    'fake bitmask error': os.path.join(BASE_PATH, 'fake_bitmask_error', 'fcs1_cleaned.lmd'),
+    'Cytek xP5': os.path.join(BASE_PATH, 'Cytek_xP5', 'Cytek_xP5.fcs'),
+    'FACS Diva': os.path.join(BASE_PATH, 'FACS_Diva', 'facs_diva_test.fcs'),
+    'guava muse': os.path.join(BASE_PATH, 'GuavaMuse','Guava Muse.fcs')
+}
 
 # The group of files below is used for checking behavior other than reading data.
 ADDITIONAL_FILE_MAPPING = {'duplicate_names': os.path.join(BASE_PATH, 'MiltenyiBiotec', 'FCS3.1',
@@ -349,7 +348,29 @@ def test_channel_naming_manual(self):
 
         self.assertListEqual(channel_names, pns_names)
         self.assertListEqual(list(data.columns.values), pns_names)
-
+
+    def test_channel_namming_FACS_Diva(self):
+        """Check that channel names correspond to manual setting."""
+        path = FILE_IDENTIFIER_TO_PATH['FACS Diva']
+        meta = parse_fcs(path, meta_data_only=True,
+                         reformat_meta=True, channel_naming='$PnN')
+        channel_names = list(meta['_channel_names_'])
+        channel_meta = meta['_channels_']
+        pnn_names = ['Time', 'FSC-A', 'FSC-W', 'SSC-A', 'FITC-A', 'PE-A', 'PerCP-A', 
+                     'PE-Cy7-A', 'PacificBlue-A', 'APC-A', 'Alexa700-A', 'APC-Cy7-A']
+        self.assertListEqual(channel_names, pnn_names)
+        self.assertListEqual(list(channel_meta["$PnN"]), pnn_names)
+
+        meta = parse_fcs(path, meta_data_only=True,
+                         reformat_meta=True, channel_naming='$PnS')
+        channel_names = list(meta['_channel_names_'])
+        channel_meta = meta['_channels_']
+        pns_names = ['Time', 'FSC-A', 'FSC-W', 'SSC-A', 'CD20', 'CD10', 'CD45',
+                      'CD34', 'Syto 41', 'CD19', 'CD38', 'APC-Cy7-A']
+        self.assertListEqual(channel_names, pns_names)
+        self.assertListEqual(list(channel_meta["$PnS"]), pns_names)
+
+
     def test_speed_of_reading_fcs_files(self):
         """Test the speed of loading a FCS files"""
         file_path = FILE_IDENTIFIER_TO_PATH['mq fcs 3.1']