"""
Analyze ELAND files
"""
-
+import collections
from glob import glob
import logging
import os
import re
import stat
import sys
+import types
-from htsworkflow.pipelines.runfolder import ElementTree, LANE_LIST
+from htsworkflow.pipelines import ElementTree, LANE_LIST
+from htsworkflow.pipelines.samplekey import SampleKey
+from htsworkflow.pipelines.genomemap import GenomeMap
from htsworkflow.util.ethelp import indent, flatten
from htsworkflow.util.opener import autoopen
XML_VERSION = 2
LANE = 'ResultLane'
- def __init__(self, pathname=None, lane_id=None, end=None, xml=None):
- self.pathname = pathname
- self._sample_name = None
+ def __init__(self, pathnames=None, sample=None, lane_id=None, end=None,
+ xml=None):
+ self.pathnames = pathnames
+ self.sample_name = sample
self.lane_id = lane_id
self.end = end
self._reads = None
"""
pass
- def _update_name(self):
- # extract the sample name
- if self.pathname is None:
- return
-
- path, name = os.path.split(self.pathname)
- split_name = name.split('_')
- self._sample_name = split_name[0]
-
- def _get_sample_name(self):
- if self._sample_name is None:
- self._update_name()
- return self._sample_name
- sample_name = property(_get_sample_name)
-
def _get_reads(self):
if self._reads is None:
self._update()
def get_elements(self):
return None
+ def __repr__(self):
+ name = []
+
+ name.append('L%s' % (self.lane_id,))
+ name.append('R%s' % (self.end,))
+ name.append('S%s' % (self.sample_name,))
+
+ return '<ResultLane(' + ",".join(name) + ')>'
+
class ElandLane(ResultLane):
"""
Process an eland result file
SCORE_QC = 1
SCORE_READ = 2
- def __init__(self, pathname=None, lane_id=None, end=None, genome_map=None, eland_type=None, xml=None):
- super(ElandLane, self).__init__(pathname, lane_id, end)
+ def __init__(self, pathnames=None, sample=None, lane_id=None, end=None,
+ genome_map=None, eland_type=None, xml=None):
+ super(ElandLane, self).__init__(pathnames, sample, lane_id, end)
self._mapped_reads = None
self._match_codes = None
- if genome_map is None:
- genome_map = {}
- self.genome_map = genome_map
+ self._reads = None
+ self.genome_map = GenomeMap(genome_map)
self.eland_type = None
if xml is not None:
self.set_elements(xml)
+ def __repr__(self):
+ name = []
+
+ name.append('L%s' % (self.lane_id,))
+ name.append('R%s' % (self.end,))
+ name.append('S%s' % (self.sample_name,))
+
+ reads = str(self._reads) if self._reads is not None else 'Uncounted'
+ return '<ElandLane(' + ",".join(name) + ' = '+ reads + ')>'
+
def _guess_eland_type(self, pathname):
if self.eland_type is None:
# attempt autodetect eland file type
Actually read the file and actually count the reads
"""
# can't do anything if we don't have a file to process
- if self.pathname is None:
+ if self.pathnames is None or len(self.pathnames) == 0:
return
- self._guess_eland_type(self.pathname)
+ pathname = self.pathnames[-1]
+ self._guess_eland_type(pathname)
- if os.stat(self.pathname)[stat.ST_SIZE] == 0:
+ if os.stat(pathname)[stat.ST_SIZE] == 0:
raise RuntimeError("Eland isn't done, try again later.")
- LOGGER.debug("summarizing results for %s" % (self.pathname))
+ LOGGER.debug("summarizing results for %s" % (pathname))
+ self._match_codes = MatchCodes()
+ self._mapped_reads = MappedReads()
+ self._reads = 0
+
+ for pathname in self.pathnames:
+ stream = autoopen(pathname, 'r')
+ if self.eland_type == ELAND_SINGLE:
+ result = self._update_eland_result(stream)
+ elif self.eland_type == ELAND_MULTI or \
+ self.eland_type == ELAND_EXTENDED:
+ result = self._update_eland_multi(stream)
+ elif self.eland_type == ELAND_EXPORT:
+ result = self._update_eland_export(stream)
+ else:
+ errmsg = "Only support single/multi/extended eland files"
+ raise NotImplementedError(errmsg)
+ stream.close()
- stream = autoopen(self.pathname, 'r')
- if self.eland_type == ELAND_SINGLE:
- result = self._update_eland_result(stream)
- elif self.eland_type == ELAND_MULTI or \
- self.eland_type == ELAND_EXTENDED:
- result = self._update_eland_multi(stream)
- elif self.eland_type == ELAND_EXPORT:
- result = self._update_eland_export(stream)
- else:
- raise NotImplementedError("Only support single/multi/extended eland files")
- self._match_codes, self._mapped_reads, self._reads = result
+ match, mapped, reads = result
+ self._match_codes += match
+ self._mapped_reads += mapped
+ self._reads += reads
def _update_eland_result(self, instream):
reads = 0
- mapped_reads = {}
+ mapped_reads = MappedReads()
+ match_codes = MatchCodes()
- match_codes = {'NM':0, 'QC':0, 'RM':0,
- 'U0':0, 'U1':0, 'U2':0,
- 'R0':0, 'R1':0, 'R2':0,
- }
for line in instream:
reads += 1
fields = line.split()
MATCH_INDEX = 2
LOCATION_INDEX = 3
reads = 0
- mapped_reads = {}
+ mapped_reads = MappedReads()
+ match_codes = MatchCodes()
- match_codes = {'NM':0, 'QC':0, 'RM':0,
- 'U0':0, 'U1':0, 'U2':0,
- 'R0':0, 'R1':0, 'R2':0,
- }
for line in instream:
reads += 1
fields = line.split()
LOCATION_INDEX = 10
DESCRIPTOR_INDEX= 13
reads = 0
- mapped_reads = {}
-
- match_codes = {'NM':0, 'QC':0, 'RM':0,
- 'U0':0, 'U1':0, 'U2':0,
- 'R0':0, 'R1':0, 'R2':0,
- }
+ mapped_reads = MappedReads()
+ match_codes = MatchCodes()
for line in instream:
reads += 1
groups = ElandLane.MATCH_COUNTS_RE.match(match)
if groups is None:
# match is not of the form [\d]+:[\d]+:[\d]+
- if match_codes.has_key(match):
+ if match in match_codes:
# match is one quality control codes QC/NM etc
match_codes[match] += 1
return ElandLane.SCORE_QC
def get_elements(self):
lane = ElementTree.Element(ElandLane.LANE,
{'version':
- unicode(ElandLane.XML_VERSION)})
+ str(ElandLane.XML_VERSION)})
sample_tag = ElementTree.SubElement(lane, SAMPLE_NAME)
sample_tag.text = self.sample_name
lane_tag = ElementTree.SubElement(lane, LANE_ID)
end_tag = ElementTree.SubElement(lane, END)
end_tag.text = str(self.end)
genome_map = ElementTree.SubElement(lane, GENOME_MAP)
- for k, v in self.genome_map.items():
+ for k, v in list(self.genome_map.items()):
item = ElementTree.SubElement(
genome_map, GENOME_ITEM,
- {'name':k, 'value':unicode(v)})
+ {'name':k, 'value':str(v)})
mapped_reads = ElementTree.SubElement(lane, MAPPED_READS)
- for k, v in self.mapped_reads.items():
+ for k, v in list(self.mapped_reads.items()):
item = ElementTree.SubElement(
mapped_reads, MAPPED_ITEM,
- {'name':k, 'value':unicode(v)})
+ {'name':k, 'value':str(v)})
match_codes = ElementTree.SubElement(lane, MATCH_CODES)
- for k, v in self.match_codes.items():
+ for k, v in list(self.match_codes.items()):
item = ElementTree.SubElement(
match_codes, MATCH_ITEM,
- {'name':k, 'value':unicode(v)})
+ {'name':k, 'value':str(v)})
reads = ElementTree.SubElement(lane, READS)
- reads.text = unicode(self.reads)
+ reads.text = str(self.reads)
return lane
for element in tree:
tag = element.tag.lower()
if tag == SAMPLE_NAME.lower():
- self._sample_name = element.text
+ self.sample_name = element.text
elif tag == LANE_ID.lower():
self.lane_id = int(element.text)
elif tag == END.lower():
else:
LOGGER.warn("ElandLane unrecognized tag %s" % (element.tag,))
+
+class MatchCodes(collections.MutableMapping):
+ """Mapping to hold match counts -
+ supports combining two match count sets together
+ """
+ def __init__(self, initializer=None):
+ self.match_codes = {'NM':0, 'QC':0, 'RM':0,
+ 'U0':0, 'U1':0, 'U2':0,
+ 'R0':0, 'R1':0, 'R2':0,
+ }
+
+ if initializer is not None:
+ if not isinstance(initializer, collections.Mapping):
+ raise ValueError("Expected dictionary like class")
+ for key in initializer:
+ if key not in self.match_codes:
+ errmsg = "Initializer can only contain: %s"
+ raise ValueError(errmsg % (",".join(list(self.match_codes.keys()))))
+ self.match_codes[key] += initializer[key]
+
+ def __iter__(self):
+ return iter(self.match_codes)
+
+ def __delitem__(self, key):
+ raise RuntimeError("delete not allowed")
+
+ def __getitem__(self, key):
+ return self.match_codes[key]
+
+ def __setitem__(self, key, value):
+ if key not in self.match_codes:
+ errmsg = "Unrecognized key, allowed values are: %s"
+ raise ValueError(errmsg % (",".join(list(self.match_codes.keys()))))
+ self.match_codes[key] = value
+
+ def __len__(self):
+ return len(self.match_codes)
+
+ def __add__(self, other):
+ if not isinstance(other, MatchCodes):
+ raise ValueError("Expected a MatchCodes, got %s", str(type(other)))
+
+ newobj = MatchCodes(self)
+ for key, value in list(other.items()):
+ newobj[key] = self.get(key, 0) + other[key]
+
+ return newobj
+
+
+class MappedReads(collections.MutableMapping):
+ """Mapping to hold mapped reads -
+ supports combining two mapped read sets together
+ """
+ def __init__(self, initializer=None):
+ self.mapped_reads = {}
+
+ if initializer is not None:
+ if not isinstance(initializer, collections.Mapping):
+ raise ValueError("Expected dictionary like class")
+ for key in initializer:
+ self[key] = self.setdefault(key, 0) + initializer[key]
+
+ def __iter__(self):
+ return iter(self.mapped_reads)
+
+ def __delitem__(self, key):
+ del self.mapped_reads[key]
+
+ def __getitem__(self, key):
+ return self.mapped_reads[key]
+
+ def __setitem__(self, key, value):
+ self.mapped_reads[key] = value
+
+ def __len__(self):
+ return len(self.mapped_reads)
+
+ def __add__(self, other):
+ if not isinstance(other, MappedReads):
+ raise ValueError("Expected a MappedReads, got %s", str(type(other)))
+
+ newobj = MappedReads(self)
+ for key in other:
+ newobj[key] = self.get(key, 0) + other[key]
+
+ return newobj
+
class SequenceLane(ResultLane):
XML_VERSION=1
LANE = 'SequenceLane'
FASTQ_TYPE = 2
SEQUENCE_DESCRIPTION = { NONE_TYPE: 'None', SCARF_TYPE: 'SCARF', FASTQ_TYPE: 'FASTQ' }
- def __init__(self, pathname=None, lane_id=None, end=None, xml=None):
+ def __init__(self, pathnames=None, sample=None, lane_id=None, end=None,
+ xml=None):
self.sequence_type = None
- super(SequenceLane, self).__init__(pathname, lane_id, end, xml)
+ super(SequenceLane, self).__init__(pathnames, sample, lane_id, end, xml)
def _guess_sequence_type(self, pathname):
"""
Actually read the file and actually count the reads
"""
# can't do anything if we don't have a file to process
- if self.pathname is None:
+ if self.pathnames is None:
return
- if os.stat(self.pathname)[stat.ST_SIZE] == 0:
+ pathname = self.pathnames[-1]
+ if os.stat(pathname)[stat.ST_SIZE] == 0:
raise RuntimeError("Sequencing isn't done, try again later.")
- self._guess_sequence_type(self.pathname)
+ self._guess_sequence_type(pathname)
- LOGGER.info("summarizing results for %s" % (self.pathname))
+ LOGGER.info("summarizing results for %s" % (pathname))
lines = 0
- f = open(self.pathname)
- for l in f.xreadlines():
+ f = open(pathname)
+ for l in f:
lines += 1
f.close()
elif self.sequence_type == SequenceLane.FASTQ_TYPE:
self._reads = lines / 4
else:
- raise NotImplementedError("This only supports scarf or fastq squence files")
+ errmsg = "This only supports scarf or fastq squence files"
+ raise NotImplementedError(errmsg)
def get_elements(self):
lane = ElementTree.Element(SequenceLane.LANE,
{'version':
- unicode(SequenceLane.XML_VERSION)})
+ str(SequenceLane.XML_VERSION)})
sample_tag = ElementTree.SubElement(lane, SAMPLE_NAME)
sample_tag.text = self.sample_name
lane_tag = ElementTree.SubElement(lane, LANE_ID)
end_tag = ElementTree.SubElement(lane, END)
end_tag.text = str(self.end)
reads = ElementTree.SubElement(lane, READS)
- reads.text = unicode(self.reads)
+ reads.text = str(self.reads)
sequence_type = ElementTree.SubElement(lane, SequenceLane.SEQUENCE_TYPE)
- sequence_type.text = unicode(SequenceLane.SEQUENCE_DESCRIPTION[self.sequence_type])
+ sequence_type.text = str(SequenceLane.SEQUENCE_DESCRIPTION[self.sequence_type])
return lane
def set_elements(self, tree):
if tree.tag != SequenceLane.LANE:
raise ValueError('Exptecting %s' % (SequenceLane.LANE,))
- lookup_sequence_type = dict([ (v,k) for k,v in SequenceLane.SEQUENCE_DESCRIPTION.items()])
+ lookup_sequence_type = dict([ (v,k) for k,v in list(SequenceLane.SEQUENCE_DESCRIPTION.items())])
for element in tree:
tag = element.tag.lower()
if tag == SAMPLE_NAME.lower():
- self._sample_name = element.text
+ self.sample_name = element.text
elif tag == LANE_ID.lower():
self.lane_id = int(element.text)
elif tag == END.lower():
else:
LOGGER.warn("SequenceLane unrecognized tag %s" % (element.tag,))
-class ELAND(object):
+class ELAND(collections.MutableMapping):
"""
Summarize information from eland files
"""
ELAND = 'ElandCollection'
LANE = 'Lane'
LANE_ID = 'id'
+ SAMPLE = 'sample'
END = 'end'
def __init__(self, xml=None):
# we need information from the gerald config.xml
- self.results = [{},{}]
+ self.results = {}
if xml is not None:
self.set_elements(xml)
- if len(self.results[0]) == 0:
- # Initialize our eland object with meaningless junk
- for l in LANE_LIST:
- self.results[0][l] = ResultLane(lane_id=l, end=0)
+ def __getitem__(self, key):
+ if not isinstance(key, SampleKey):
+ raise ValueError("Key must be a %s" % (str(type(SampleKey))))
+ return self.results[key]
+
+ def __setitem__(self, key, value):
+ if not isinstance(key, SampleKey):
+ raise ValueError("Key must be a %s" % (str(type(SampleKey))))
+ self.results[key] = value
+
+ def __delitem__(self, key):
+ del self.result[key]
+
+ def __iter__(self):
+ keys = iter(self.results.keys())
+ for k in sorted(keys):
+ yield k
+
+ def __len__(self):
+ return len(self.results)
+ def find_keys(self, search):
+ """Return results that match key"""
+ if not isinstance(search, SampleKey):
+ raise ValueError("Key must be a %s" % (str(type(SampleKey))))
+ if not search.iswild:
+ yield self[search]
+ for key in list(self.keys()):
+ if key.matches(search): yield key
def get_elements(self):
root = ElementTree.Element(ELAND.ELAND,
- {'version': unicode(ELAND.XML_VERSION)})
- for end in range(len(self.results)):
- end_results = self.results[end]
- for lane_id, lane in end_results.items():
- eland_lane = lane.get_elements()
- if eland_lane is not None:
- eland_lane.attrib[ELAND.END] = unicode (end)
- eland_lane.attrib[ELAND.LANE_ID] = unicode(lane_id)
- root.append(eland_lane)
+ {'version': str(ELAND.XML_VERSION)})
+
+ for key in self:
+ eland_lane = self[key].get_elements()
+ eland_lane.attrib[ELAND.END] = str(self[key].end-1)
+ eland_lane.attrib[ELAND.LANE_ID] = str(self[key].lane_id)
+ eland_lane.attrib[ELAND.SAMPLE] = str(self[key].sample_name)
+ root.append(eland_lane)
+ return root
return root
def set_elements(self, tree):
for element in list(tree):
lane_id = int(element.attrib[ELAND.LANE_ID])
end = int(element.attrib.get(ELAND.END, 0))
+ sample = element.attrib.get(ELAND.SAMPLE, 's')
if element.tag.lower() == ElandLane.LANE.lower():
lane = ElandLane(xml=element)
elif element.tag.lower() == SequenceLane.LANE.lower():
lane = SequenceLane(xml=element)
- self.results[end][lane_id] = lane
-
-def check_for_eland_file(basedir, pattern, lane_id, end):
- if end is None:
- full_lane_id = lane_id
- else:
- full_lane_id = "%d_%d" % ( lane_id, end )
-
- basename = pattern % (full_lane_id,)
- LOGGER.debug("Eland pattern: %s" %(basename,))
- pathname = os.path.join(basedir, basename)
- if os.path.exists(pathname):
- LOGGER.info('found eland file in %s' % (pathname,))
- return pathname
- else:
- return None
-
-def update_result_with_eland(gerald, results, lane_id, end, pathname, genome_maps):
- # yes the lane_id is also being computed in ElandLane._update
- # I didn't want to clutter up my constructor
- # but I needed to persist the sample_name/lane_id for
- # runfolder summary_report
- path, name = os.path.split(pathname)
- LOGGER.info("Adding eland file %s" %(name,))
- # split_name = name.split('_')
- # lane_id = int(split_name[1])
-
- genome_map = {}
- if genome_maps is not None:
- genome_map = genome_maps[lane_id]
- elif gerald is not None:
- genome_dir = gerald.lanes[lane_id].eland_genome
- if genome_dir is not None:
- genome_map = build_genome_fasta_map(genome_dir)
-
- lane = ElandLane(pathname, lane_id, end, genome_map)
-
- if end is None:
- effective_end = 0
- else:
- effective_end = end - 1
-
- results[effective_end][lane_id] = lane
-
-def update_result_with_sequence(gerald, results, lane_id, end, pathname):
- result = SequenceLane(pathname, lane_id, end)
-
- if end is None:
- effective_end = 0
- else:
- effective_end = end - 1
-
- results[effective_end][lane_id] = result
+ key = SampleKey(lane=lane_id, read=end+1, sample=sample)
+ self.results[key] = lane
-def eland(gerald_dir, gerald=None, genome_maps=None):
- e = ELAND()
+ def update_result_with_eland(self, gerald, key, pathnames,
+ genome_maps):
+ # yes the lane_id is also being computed in ElandLane._update
+ # I didn't want to clutter up my constructor
+ # but I needed to persist the sample_name/lane_id for
+ # runfolder summary_report
+ names = [ os.path.split(p)[1] for p in pathnames]
+ LOGGER.info("Adding eland files %s" %(",".join(names),))
+ basedir = os.path.split(pathnames[0])[0]
+ gs_template = "{0}_*_L{1:03}_genomesize.xml"
+ genomesize = glob(
+ os.path.join(basedir,
+ gs_template.format(key.sample, key.lane)))
+
+
+ genome_map = GenomeMap()
+ if genome_maps is not None:
+ genome_map = GenomeMap(genome_maps[key.lane])
+ elif len(genomesize) > 0:
+ LOGGER.info("Found {0}".format(genomesize))
+ genome_map.parse_genomesize(genomesize[0])
+ elif gerald is not None:
+ genome_dir = gerald.lanes[key].eland_genome
+ if genome_dir is not None:
+ genome_map.scan_genome_dir(genome_dir)
+
+ lane = ElandLane(pathnames, key.sample, key.lane, key.read, genome_map)
+
+ self.results[key] = lane
- lane_ids = range(1,9)
- ends = [None, 1, 2]
-
- basedirs = [gerald_dir]
-
- # if there is a basedir/Temp change basedir to point to the temp
- # directory, as 1.1rc1 moves most of the files we've historically
- # cared about to that subdirectory.
- # we should look into what the official 'result' files are.
- # and 1.3 moves them back
- basedir_temp = os.path.join(gerald_dir, 'Temp')
- if os.path.isdir(basedir_temp):
- basedirs.append(basedir_temp)
-
-
- # the order in patterns determines the preference for what
- # will be found.
- MAPPED_ELAND = 0
- SEQUENCE = 1
- patterns = [('s_%s_eland_result.txt', MAPPED_ELAND),
- ('s_%s_eland_result.txt.bz2', MAPPED_ELAND),
- ('s_%s_eland_result.txt.gz', MAPPED_ELAND),
- ('s_%s_eland_extended.txt', MAPPED_ELAND),
- ('s_%s_eland_extended.txt.bz2', MAPPED_ELAND),
- ('s_%s_eland_extended.txt.gz', MAPPED_ELAND),
- ('s_%s_eland_multi.txt', MAPPED_ELAND),
- ('s_%s_eland_multi.txt.bz2', MAPPED_ELAND),
- ('s_%s_eland_multi.txt.gz', MAPPED_ELAND),
- ('s_%s_export.txt', MAPPED_ELAND),
- ('s_%s_export.txt.bz2', MAPPED_ELAND),
- ('s_%s_export.txt.gz', MAPPED_ELAND),
- ('s_%s_sequence.txt', SEQUENCE),]
-
- for basedir in basedirs:
- for end in ends:
- for lane_id in lane_ids:
- for p in patterns:
- pathname = check_for_eland_file(basedir, p[0], lane_id, end)
- if pathname is not None:
- if p[1] == MAPPED_ELAND:
- update_result_with_eland(gerald, e.results, lane_id, end, pathname, genome_maps)
- elif p[1] == SEQUENCE:
- update_result_with_sequence(gerald, e.results, lane_id, end, pathname)
- break
- else:
- LOGGER.debug("No eland file found in %s for lane %s and end %s" %(basedir, lane_id, end))
- continue
+ def update_result_with_sequence(self, gerald, key, pathnames,
+ genome_maps=None):
+ self.results[key] = SequenceLane(pathnames,
+ key.sample, key.lane, key.read)
+
+def eland(gerald_dir, gerald=None, genome_maps=None):
+ e = ELAND()
+ eland_files = ElandMatches(e)
+ # collect
+ for path, dirnames, filenames in os.walk(gerald_dir):
+ for filename in filenames:
+ pathname = os.path.abspath(os.path.join(path, filename))
+ eland_files.add(pathname)
+ for key in eland_files:
+ eland_files.count(key, gerald, genome_maps)
return e
-def build_genome_fasta_map(genome_dir):
- # build fasta to fasta file map
- LOGGER.info("Building genome map")
- genome = genome_dir.split(os.path.sep)[-1]
- fasta_map = {}
- for vld_file in glob(os.path.join(genome_dir, '*.vld')):
- is_link = False
- if os.path.islink(vld_file):
- is_link = True
- vld_file = os.path.realpath(vld_file)
- path, vld_name = os.path.split(vld_file)
- name, ext = os.path.splitext(vld_name)
- if is_link:
- fasta_map[name] = name
- else:
- fasta_map[name] = os.path.join(genome, name)
- return fasta_map
+
+class ElandMatches(collections.MutableMapping):
+ def __init__(self, eland_container):
+ # the order in patterns determines the preference for what
+ # will be found.
+ self.eland_container = eland_container
+ MAPPED = eland_container.update_result_with_eland
+ SEQUENCE = eland_container.update_result_with_sequence
+
+ sample = '(?P<sample>[^_]+)'
+ hiIndex = '_(?P<index>(NoIndex|[AGCT])+)'
+ hiLane = '_L(?P<lane>[\d]+)'
+ gaLane = '_(?P<lane>[\d]+)'
+ hiRead = '_R(?P<read>[\d]+)'
+ gaRead = '(_(?P<read>[\d])+)?'
+ part = '_(?P<part>[\d]+)'
+ ext = '(?P<extention>(\.bz2|\.gz)?)'
+
+ hiPrefix = sample + hiIndex + hiLane + hiRead + part
+ gaPrefix = sample + gaLane + gaRead
+ P = collections.namedtuple('Patterns', 'pattern counter priority')
+ self.patterns = [
+ P(hiPrefix +'_export.txt' + ext, MAPPED, 6),
+ P(gaPrefix + '_eland_result.txt' + ext, MAPPED, 5),
+ P(gaPrefix + '_eland_extended.txt' + ext, MAPPED, 4),
+ P(gaPrefix + '_eland_multi.txt' + ext, MAPPED, 3),
+ P(gaPrefix + '_export.txt' + ext, MAPPED, 2),
+ P(gaPrefix + '_sequence.txt' + ext, SEQUENCE, 1),
+ ]
+ self.file_sets = {}
+ self.file_priority = {}
+ self.file_counter = {}
+
+ def add(self, pathname):
+ """Add pathname to our set of files
+ """
+ path, filename = os.path.split(pathname)
+
+ for pattern, counter, priority in self.patterns:
+ rematch = re.match(pattern, filename)
+ if rematch is not None:
+ m = ElandMatch(pathname, counter, **rematch.groupdict())
+ key = m.make_samplekey()
+ old_priority = self.file_priority.get(key, 0)
+ if priority > old_priority:
+ self.file_sets[key] = set((m,))
+ self.file_counter[key] = counter
+ self.file_priority[key] = priority
+ elif priority == old_priority:
+ self.file_sets[key].add(m)
+
+ def count(self, key, gerald=None, genome_maps=None):
+ #previous sig: gerald, e.results, lane_id, end, pathnames, genome_maps
+ counter = self.file_counter[key]
+ file_set = self.file_sets[key]
+ filenames = [ f.filename for f in file_set ]
+ return counter(gerald, key,
+ filenames, genome_maps)
+
+ def __iter__(self):
+ return iter(self.file_sets)
+
+ def __len__(self):
+ return len(self.file_sets)
+
+ def __getitem__(self, key):
+ return self.file_sets[key]
+
+ def __setitem__(self, key, value):
+ if not isintance(value, set):
+ raise ValueError("Expected set for value")
+ self.file_sets[key] = value
+
+ def __delitem__(self, key):
+ del self.file_sets[key]
+
+class ElandMatch(object):
+ def __init__(self, pathname, counter,
+ lane=None, read=None, extension=None,
+ sample=None, index=None, part=None, **kwargs):
+ self.filename = pathname
+ self.counter = counter
+ self._lane = lane
+ self._read = read
+ self.extension = extension
+ self.sample = sample
+ self.index = index
+ self._part = part
+ LOGGER.info("Found %s: L%s R%s Samp%s" % (
+ self.filename, self._lane, self._read, self.sample))
+
+ def make_samplekey(self):
+ read = self._read if self._read is not None else 1
+ return SampleKey(lane=self.lane, read=read, sample=self.sample)
+
+ def _get_lane(self):
+ if self._lane is not None:
+ return int(self._lane)
+ return self._lane
+ lane = property(_get_lane)
+
+ def _get_read(self):
+ if self._read is not None:
+ return int(self._read)
+ return self._read
+ read = property(_get_read)
+
+ def _get_part(self):
+ if self._part is not None:
+ return int(self._part)
+ return self._part
+ part = property(_get_part)
+
+ def __repr__(self):
+ name = []
+ if self.sample is not None: name.append(self.sample)
+ if self._lane is not None: name.append('L%s' % (self.lane,))
+ if self._read is not None: name.append('R%s' % (self.read,))
+ if self._part is not None: name.append('P%s' % (self.part,))
+ return '<ElandMatch(' + "_".join(name) + ')>'
def extract_eland_sequence(instream, outstream, start, end):
from optparse import OptionParser
parser = OptionParser("%prog: <gerald dir>+")
opts, args = parser.parse_args(cmdline)
- LOGGER.basicConfig(level=logging.DEBUG)
+ logging.basicConfig(level=logging.DEBUG)
for a in args:
LOGGER.info("Starting scan of %s" % (a,))
e = eland(a)
- print e.get_elements()
-
+ print(ElementTree.tostring(e.get_elements()))
return
projects / htsworkflow.git / blobdiff