Merge changing lane_number to string and sequence finding code changes

[htsworkflow.git] / htsworkflow / pipelines / sequences.py

diff --git a/htsworkflow/pipelines/sequences.py b/htsworkflow/pipelines/sequences.py
index 5baf1b5022a948b1022925737bc79cd3e425aafd..acd100575744c94984d7b4b1ab5074188145844e 100644 (file)
--- a/htsworkflow/pipelines/sequences.py
+++ b/htsworkflow/pipelines/sequences.py
@@ -164,7 +164,7 @@ class SequenceFile(object):
         # a bit unreliable... assumes filesystem is encoded in utf-8
         path = os.path.abspath(self.path.encode('utf-8'))
         fileNode = RDF.Node(RDF.Uri('file://' + path))
-        add(model, fileNode, rdfNS['type'], libNS['raw_file'])
+        add(model, fileNode, rdfNS['type'], libNS['IlluminaResult'])
         add_lit(model, fileNode, libNS['flowcell_id'], self.flowcell)
         add_lit(model, fileNode, libNS['lane_number'], self.lane)
         if self.read is not None:
@@ -177,7 +177,7 @@ class SequenceFile(object):
         add_lit(model, fileNode, libNS['split_id'], self.split)
         add_lit(model, fileNode, libNS['cycle'], self.cycle)
         add_lit(model, fileNode, libNS['passed_filter'], self.pf)
-        add(model, fileNode, rdfNS['type'], libNS[self.filetype])
+        add(model, fileNode, libNS['file_type'], libNS[self.filetype])
 
         if base_url is not None:
             flowcell = RDF.Node(RDF.Uri("{base}/flowcell/{flowcell}/".format(
@@ -215,18 +215,15 @@ class SequenceFile(object):
 
         if not isinstance(seq_id, RDF.Node):
             seq_id = RDF.Node(RDF.Uri(seq_id))
-        seqTypesStmt = RDF.Statement(seq_id, rdfNS['type'], None)
-        seqTypes = model.find_statements(seqTypesStmt)
-        isSequenceFile = False
-        for s in seqTypes:
-            if s.object == libNS['raw_file']:
-                isSequenceFile = True
-            else:
-                seq_type = stripNamespace(libNS, s.object)
-
-        if not isSequenceFile:
+        result_statement = RDF.Statement(seq_id,
+                                         rdfNS['type'],
+                                         libNS['IlluminaResult'])
+        if not model.contains_statement(result_statement):
             raise KeyError(u"%s not found" % (unicode(seq_id),))
 
+        seq_type_node = model.get_target(seq_id, libNS['file_type'])
+        seq_type = stripNamespace(libNS, seq_type_node)
+
         path = urlparse(str(seq_id.uri)).path
         flowcellNode = get_one(seq_id, libNS['flowcell'])
         flowcell = get_one(seq_id, libNS['flowcell_id'])
@@ -278,7 +275,7 @@ def parse_srf(path, filename):
     basename, ext = os.path.splitext(filename)
     records = basename.split('_')
     flowcell = records[4]
-    lane = int(records[5][0])
+    lane = records[5][0]
     fullpath = os.path.join(path, filename)
 
     if flowcell_dir != flowcell:
@@ -293,7 +290,7 @@ def parse_qseq(path, filename):
     records = basename.split('_')
     fullpath = os.path.join(path, filename)
     flowcell = records[4]
-    lane = int(records[5][1])
+    lane = records[5][1]
     read = int(records[6][1])
 
     if flowcell_dir != flowcell:
@@ -312,7 +309,7 @@ def parse_fastq(path, filename):
     if project is not None:
         # demultiplexed sample!
         flowcell = flowcell_dir
-        lane = int(records[2][-1])
+        lane = records[2][-1]
         read = int(records[3][-1])
         pf = True # as I understand it hiseq runs toss the ones that fail filter
         index = records[1]
@@ -321,7 +318,7 @@ def parse_fastq(path, filename):
         sequence_type = 'split_fastq'
     else:
         flowcell = records[4]
-        lane = int(records[5][1])
+        lane = records[5][1]
         read = int(records[6][1])
         pf = parse_fastq_pf_flag(records)
         index = None
@@ -365,7 +362,7 @@ def parse_eland(path, filename, eland_match=None):
     fullpath = os.path.join(path, filename)
     flowcell, start, stop, project = get_flowcell_cycle(path)
     if eland_match.group('lane'):
-        lane = int(eland_match.group('lane'))
+        lane = eland_match.group('lane')
     else:
         lane = None
     if eland_match.group('read'):
@@ -418,17 +415,18 @@ def update_model_sequence_library(model, base_url):
     """Find sequence objects and add library information if its missing
     """
     file_body = """
-    prefix libNS: <http://jumpgate.caltech.edu/wiki/LibraryOntology#>
+    prefix libns: <http://jumpgate.caltech.edu/wiki/LibraryOntology#>
     select ?filenode ?flowcell_id ?lane_id ?library_id ?flowcell ?library
     where {
-       ?filenode a libNS:raw_file ;
-                 libNS:flowcell_id ?flowcell_id ;
-                 libNS:lane_number ?lane_id .
-       OPTIONAL { ?filenode libNS:flowcell ?flowcell . }
-       OPTIONAL { ?filenode libNS:library ?library .}
-       OPTIONAL { ?filenode libNS:library_id ?library_id .}
+       ?filenode a libns:IlluminaResult ;
+                 libns:flowcell_id ?flowcell_id ;
+                 libns:lane_number ?lane_id .
+       OPTIONAL { ?filenode libns:flowcell ?flowcell . }
+       OPTIONAL { ?filenode libns:library ?library .}
+       OPTIONAL { ?filenode libns:library_id ?library_id .}
     }
     """
+    LOGGER.debug("update_model_sequence_library query %s", file_body)
     file_query = RDF.SPARQLQuery(file_body)
     files = file_query.execute(model)
 
@@ -436,9 +434,13 @@ def update_model_sequence_library(model, base_url):
     flowcellNS = RDF.NS(urljoin(base_url, 'flowcell/'))
     for f in files:
         filenode = f['filenode']
+        LOGGER.debug("Updating file node %s", str(filenode))
         lane_id = fromTypedNode(f['lane_id'])
         if f['flowcell'] is None:
             flowcell = flowcellNS[str(f['flowcell_id'])+'/']
+            LOGGER.debug("Adding file (%s) to flowcell (%s) link",
+                         str(filenode),
+                         str(flowcell))
             model.add_statement(
                 RDF.Statement(filenode, libNS['flowcell'], flowcell))
         else:
@@ -452,6 +454,9 @@ def update_model_sequence_library(model, base_url):
                                                    flowcell,
                                                    lane_id)
                 library_id = toTypedNode(simplify_uri(library))
+                LOGGER.debug("Adding file (%s) to library (%s) link",
+                             str(filenode),
+                             str(library))
                 model.add_statement(
                     RDF.Statement(filenode, libNS['library_id'], library_id))
             if library is not None:
@@ -465,16 +470,16 @@ def guess_library_from_model(model, base_url, flowcell, lane_id):
     flowcellNode = RDF.Node(flowcell)
     flowcell = str(flowcell.uri)
     lane_body = """
-    prefix libNS: <http://jumpgate.caltech.edu/wiki/LibraryOntology#>
+    prefix libns: <http://jumpgate.caltech.edu/wiki/LibraryOntology#>
     prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
     prefix xsd: <http://www.w3.org/2001/XMLSchema#>
 
     select ?library ?lane
     where {{
-      <{flowcell}> libNS:has_lane ?lane ;
-                   a libNS:illumina_flowcell .
-      ?lane libNS:lane_number {lane_id} ;
-            libNS:library ?library .
+      <{flowcell}> libns:has_lane ?lane ;
+                   a libns:IlluminaFlowcell .
+      ?lane libns:lane_number "{lane_id}" ;
+            libns:library ?library .
     }}
     """
     lane_body = lane_body.format(flowcell=flowcell, lane_id=lane_id)