Merge pull request #54 from MatildaAslin/issue_53

Read length equal to number of cycles

Author: johandahlberg

checkQC/__init__.py

@@ -1,2 +1,2 @@
 
-__version__ = "1.3.0"
+__version__ = "1.4.0"

checkQC/config.py

@@ -1,6 +1,7 @@
 
 from pkg_resources import Requirement, resource_filename
 import logging
+from checkQC.exceptions import ConfigEntryMissing
 
 import yaml
 
@@ -84,19 +85,23 @@ def _get_matching_handler(self, instrument_and_reagent_type, read_length):
         :param instrument_and_reagent_type: the instrument and run type, e.g. 'hiseq2500_rapidhighoutput_v4'
         :param read_length: either as a range, e.g. '50-70' or a single value, e.g. '50'
         :returns: A dict corresponding to the handler config
+        :raises: ConfigEntryMissing if instrument, reagent type and read length detected is missing from config
         """
         config_read_lengths = list(map(str, self._config[instrument_and_reagent_type].keys()))
         for config_read_length in config_read_lengths:
             if "-" in config_read_length:
                 split_read_length = config_read_length.split("-")
                 low_break = int(split_read_length[0])
                 high_break = int(split_read_length[1])
-                if low_break < int(read_length) <= high_break:
+                if low_break <= int(read_length) <= high_break:
                     return self._config[instrument_and_reagent_type][config_read_length]["handlers"]
             else:
                 if int(read_length) == int(config_read_length):
                     return self._config[instrument_and_reagent_type][int(config_read_length)]["handlers"]
-        raise KeyError
+        raise ConfigEntryMissing("Could not find a config entry for instrument '{}' "
+                  "with read length '{}'. Please check the provided config "
+                  "file ".format(instrument_and_reagent_type,
+                                 read_length))
 
     def _add_default_config(self, current_handler_config):
         """
@@ -121,17 +126,19 @@ def get_handler_configs(self, instrument_and_reagent_type, read_length):
         :param read_length: give the read length either as str or int
         :returns: the corresponding handler configuration(s)
         """
+
         try:
             handler_config = self._get_matching_handler(instrument_and_reagent_type, read_length)
             handler_config_with_defaults = self._add_default_config(handler_config)
             return handler_config_with_defaults
-        except KeyError as e:
+        except ConfigEntryMissing as e:
             log.error("Could not find a config entry for instrument '{}' "
                       "with read length '{}'. Please check the provided config "
                       "file ".format(instrument_and_reagent_type,
                                      read_length))
             raise e
 
+
     def __getitem__(self, key):
         return self._config[key]
 

checkQC/default_config/config.yaml

@@ -1,21 +1,6 @@
 
 
-# Usage instruction for config
-# -----------------------------
-#  - Please note that intervals for read lengths are specified as: min < x <= max (i.e. upper inclusive, lower exclusive)
-#  - All other intervals are exclusive.
-#  - Values that are specified under each handler, e.g.
-#
-#      - name: ClusterPFHandler
-#        warning: 180 # Millons of clusters
-#        error: unknown
-#
-#    are specific to that partiular handler, but in general any value can be substituted with "unknown", in which case
-#    this will not be evaluated.
-#
-#  - Handlers specified under "default_handlers" will be run regardless of instrument type. For all other cases
-#    it is possible to specify handlers per instrument and read length interval.
-#
+# For information about config usage, see http://checkqc.readthedocs.io/en/latest/#configuration-file
 
 # Use this section to provide configuration options to the parsers
 parser_configurations:
@@ -30,7 +15,7 @@ default_handlers:
       error: 10
 
 hiseq2500_rapidhighoutput_v4:
-  50-70:
+  51-71:
     handlers:
       - name: ClusterPFHandler
         warning: 180 # Millons of clusters
@@ -45,7 +30,7 @@ hiseq2500_rapidhighoutput_v4:
       - name: ReadsPerSampleHandler
         warning: unknown
         error: 90 # 50 % of threshold for clusters pass filter
-  100-110:
+  101-111:
     handlers:
       - name: ClusterPFHandler
         warning: 180 # Millons of clusters
@@ -60,7 +45,7 @@ hiseq2500_rapidhighoutput_v4:
       - name: ReadsPerSampleHandler
         warning: unknown
         error: 90 # 50 % of threshold for clusters pass filter
-  120-130:
+  121-131:
     handlers:
       - name: ClusterPFHandler
         warning: 180 # Millons of clusters
@@ -77,7 +62,7 @@ hiseq2500_rapidhighoutput_v4:
         error: 90 # 50 % of threshold for clusters pass filter
 
 hiseq2500_rapidrun_v2:
-  50:
+  51:
     handlers:
       - name: ClusterPFHandler
         warning: 110 # Millons of clusters
@@ -92,7 +77,7 @@ hiseq2500_rapidrun_v2:
       - name: ReadsPerSampleHandler
         warning: unknown
         error: 55 # 50 % of threshold for clusters pass filter
-  100:
+  101:
     handlers:
       - name: ClusterPFHandler
         warning: 110 # Millons of clusters
@@ -107,7 +92,7 @@ hiseq2500_rapidrun_v2:
       - name: ReadsPerSampleHandler
         warning: unknown
         error: 55 # 50 % of threshold for clusters pass filter
-  250:
+  251:
     handlers:
       - name: ClusterPFHandler
         warning: 110 # Millons of clusters
@@ -124,7 +109,7 @@ hiseq2500_rapidrun_v2:
         error: 55 # 50 % of threshold for clusters pass filter
 
 hiseqx_v2:
-  150:
+  151:
     handlers:
       - name: ClusterPFHandler
         warning: 400 # Millons of clusters
@@ -141,7 +126,7 @@ hiseqx_v2:
         error: 200 # 50 % of threshold for clusters pass filter
 
 novaseq_v1:
-  150:
+  151:
     handlers:
       - name: ClusterPFHandler
         warning: 1400 # Millons of clusters
@@ -158,7 +143,7 @@ novaseq_v1:
         error: 700 # 50 % of threshold for clusters pass filter
 
 miseq_v2:
-  50:
+  51:
     handlers:
       - name: ClusterPFHandler
         warning: 10 # Millons of clusters
@@ -173,7 +158,7 @@ miseq_v2:
       - name: ReadsPerSampleHandler
         warning: unknown
         error: 5 # 50 % of threshold for clusters pass filter
-  150:
+  151:
     handlers:
       - name: ClusterPFHandler
         warning: 10 # Millons of clusters
@@ -188,7 +173,7 @@ miseq_v2:
       - name: ReadsPerSampleHandler
         warning: unknown
         error: 5 # 50 % of threshold for clusters pass filter
-  250:
+  251:
     handlers:
       - name: ClusterPFHandler
         warning: 10 # Millons of clusters
@@ -205,7 +190,7 @@ miseq_v2:
         error: 5 # 50 % of threshold for clusters pass filter
 
 miseq_v3:
-  75:
+  76:
     handlers:
       - name: ClusterPFHandler
         warning: 18 # Millons of clusters
@@ -220,7 +205,7 @@ miseq_v3:
       - name: ReadsPerSampleHandler
         warning: unknown
         error: 9 # 50 % of threshold for clusters pass filter
-  300:
+  301:
     handlers:
       - name: ClusterPFHandler
         warning: 18 # Millons of clusters

checkQC/exceptions.py

@@ -39,3 +39,7 @@ class QCHandlerNotFound(CheckQCException):
 
 class ConfigurationError(CheckQCException):
     pass
+
+
+class ConfigEntryMissing(CheckQCException):
+    pass

checkQC/run_type_recognizer.py

@@ -215,10 +215,7 @@ def read_length(self):
         read_lengths = []
         for read in reads:
             if not read['@IsIndexedRead'] == 'Y':
-                # The -1 is necessary for the number of cycles to correspond to the
-                # way it is specified in the docs. I.e. read length 300 in the docs
-                # means 301 cycles were run...
-                read_lengths.append(int(read['@NumCycles']) - 1)
+                read_lengths.append(int(read['@NumCycles']))
 
         if len(read_lengths) < 1:
             raise RunModeUnknown("Found no NumCycles in RunInfo.xml, could not determine read length")

docs/index.rst

@@ -18,7 +18,7 @@ Instrument types supported in checkQC are the following:
  - HiSeq2500
  - MiSeq
  - NovaSeq
-     
+
 Install instructions
 --------------------
 CheckQC **requires Python 3.5** (or higher to run). Furthermore, right now the Illumina Interop
@@ -49,7 +49,7 @@ your own custom file, you can do so by adding a path to the config like this:
 
 When CheckQC starts and no path to the config file is specified it will give you
 the path to where the default file is located on your system, if you want a template
-that you can customize according to your own needs.
+that you can customize according to your own needs. See section `Configuration file`_ for more information.
 
 When you run CheckQC you can expect to see output similar to this:
 
@@ -199,6 +199,28 @@ In this example we use the python json tool to pretty print the json output:
       }
   }
 
+Configuration file
+------------------
+
+ - The location of the default config is printed when running CheckQC without the `--config` flag.
+   It can be used to as a template when making a customized config.
+
+ - Read length is defined as the number of cycles run for a read.
+
+ - All intervals for read lengths are specified as: min <= x <= max (i.e. upper inclusive, lower inclusive).
+
+ - All other intervals are exclusive.
+
+ - Values that are specified under each handler are specific to that particular handler, but in general any value
+   can be substituted with "unknown", in which case this will not be evaluated.
+
+ - Handlers specified under "default_handlers" will be run regardless of instrument type. For all other cases it
+   is possible to specify handlers per instrument and read length interval.
+
+ - Apart from QC thresholds, the config also contains parser configurations, where parser specific variables can be set.
+   The Stats.json parser has a bcl2fastq_output_path variable, that can be set to specify where bcl2fastq output is located
+   relative to the runfolder. Default value is "Data/Intensities/BaseCalls".
+
 Running CheckQC as a webservice
 -------------------------------
 

tests/test_config.py

@@ -3,6 +3,8 @@
 
 from checkQC.config import Config, ConfigFactory
 
+from checkQC.exceptions import ConfigEntryMissing
+
 
 class TestConfig(unittest.TestCase):
 
@@ -33,7 +35,7 @@ def test_interval_match(self):
         self.assertListEqual(handlers, [self.second_handler, self.default_handler, self.first_handler])
 
     def test_no_match(self):
-        with self.assertRaises(KeyError):
+        with self.assertRaises(ConfigEntryMissing):
             self.config.get_handler_configs('miseq_v3', 999)
 
     def test_call_with_str(self):

tests/test_run_type_recognizer.py

@@ -19,7 +19,7 @@ def test_instrument_type(self):
         self.assertEqual(expected, actual.name())
 
     def test_read_length(self):
-        expected = "50"
+        expected = "51"
         actual = self.runtype_recognizer.read_length()
         self.assertEqual(expected, actual)