From: Diane Trout <diane@caltech.edu>
Date: Tue, 6 Oct 2009 18:34:36 +0000 (+0000)
Subject: Use eland_extended or eland_pair instead of eland for ANALYSIS type
X-Git-Tag: 0.3.2~10
X-Git-Url: http://woldlab.caltech.edu/gitweb/?p=htsworkflow.git;a=commitdiff_plain;h=c5ece5c8ba4ae17d9a2fdaaebadf85453aa130c6

Use eland_extended or eland_pair instead of eland for ANALYSIS type
this required splitting up my analysis suffix for sequencing and
aligning code.

Also forcing lanes that have no available genome to be sequencing
changed part of a retrive_config test case.
---

diff --git a/htsworkflow/pipelines/retrieve_config.py b/htsworkflow/pipelines/retrieve_config.py
index 1a5e136..136bb5a 100644
--- a/htsworkflow/pipelines/retrieve_config.py
+++ b/htsworkflow/pipelines/retrieve_config.py
@@ -116,13 +116,15 @@ def format_gerald_config(options, flowcell_info, genome_map):
     Generate a GERALD config file
     """
     # so we can add nothing or _pair if we're a paired end run
-    run_type_suffix = { False: "", True: "_pair" }
+    eland_analysis_suffix = { False: "_extended", True: "_pair" }
+    sequence_analysis_suffix = { False: "", True: "_pair" }
 
     # it's convienent to have helpful information describing the flowcell
     # in the config file... things like which lane is which library.
     config = [format_gerald_header(flowcell_info)]
 
-    analysis_suffix = run_type_suffix[flowcell_info['paired_end']]
+    analysis_suffix = eland_analysis_suffix[flowcell_info['paired_end']]
+    sequence_suffix = sequence_analysis_suffix[flowcell_info['paired_end']]
     lane_groups = group_lane_parameters(flowcell_info)
     for lane_index, lane_numbers in lane_groups.items():
         # lane_index is return value of group_lane_parameters
@@ -137,11 +139,11 @@ def format_gerald_config(options, flowcell_info, genome_map):
             logging.warning(no_genome_msg % (lane_numbers, species))
             is_sequencing = True
             
-        if not is_sequencing:
+        if is_sequencing:
+            config += ['%s:ANALYSIS sequence%s' % (lane_prefix, analysis_suffix)]
+        else:
             config += ['%s:ANALYSIS eland%s' % (lane_prefix, analysis_suffix)]
             config += ['%s:ELAND_GENOME %s' % (lane_prefix, species_path) ]
-        else:
-            config += ['%s:ANALYSIS sequence%s' % (lane_prefix, analysis_suffix)]
         #config += ['%s:READ_LENGTH %s' % ( lane_prefix, read_length ) ]
         config += ['%s:USE_BASES Y%s' % ( lane_prefix, read_length ) ]
 
diff --git a/htsworkflow/pipelines/test/test_retrive_config.py b/htsworkflow/pipelines/test/test_retrive_config.py
index dd8f30e..9847ffd 100644
--- a/htsworkflow/pipelines/test/test_retrive_config.py
+++ b/htsworkflow/pipelines/test/test_retrive_config.py
@@ -35,8 +35,10 @@ class RetrieveTestCases(TestCase):
         human = [ line for line in config_lines if re.search('hg18', line) ]
         self.failUnlessEqual(len(human), 1)
         self.failUnlessEqual(human[0], '345678:ELAND_GENOME /tmp/hg18')
-        unknown = [ line for line in config_lines if re.search('Unknown', line) ]
-        self.failUnlessEqual(len(unknown), 2)
+        # we changed the api to force unknown genomes to be sequencing
+        sequencing = [ line for line in config_lines if re.search('sequence_pair', line) ]
+        self.failUnlessEqual(len(sequencing), 2)
+