def merge(self):
"""merge runs from parallel computations.
returns true if merging was succecss.
"""
if self.isComplete(): return
infiles = glob.glob("%s*" % self.mFilenameProfile)
# remove suffixes
infiles = list(
set([x[:-4] for x in infiles if x != self.mFilenameProfile]))
infiles.sort()
last_nid = None
found = set()
ninput, noutput, nfound, nunknown, nduplicate = 0, 0, 0, 0, 0
tokens = set(self.mFasta.keys())
self.mProfileLibrary = ProfileLibrary.ProfileLibrary(
self.mFilenameProfile, "w")
for filename in infiles:
infile = ProfileLibrary.ProfileLibrary(filename, "r")
for nid, profile in infile.iteritems_sorted():
ninput += 1
if nid in found:
nduplicates += 1
self.warn("duplicate nid: %i in file %s" % (nid, filename))
if nid not in tokens:
nunknown += 1
self.warn("unknown nid: %i in file %s" % (nid, filename))
found.add(nid)
nfound += 1
self.mProfileLibrary.add(nid, profile)
noutput += 1
missing = tokens.difference(found)
if len(missing) > 0:
self.warn("the following nids were missing: %s" % str(missing))
self.info("adding %i missing nids" % len(missing))
for nid in missing:
self.applyMethod(AddaIO.NeighboursRecord(nid, []))
self.info( "merging: parts=%i, ninput=%i, noutput=%i, nfound=%i, nmissing=%i, nduplicate=%i, nunknown=%i" %\
(len(infiles), ninput, noutput, nfound, len(missing), nduplicate, nunknown ) )
self.info("deleting %i parts" % len(infiles))
for infile in infiles:
fn, fi = ProfileLibrary.getFileNames(infile)
os.remove(fn)
os.remove(fi)
return len(missing) == 0 and nduplicate == 0 and nunknown == 0
def finish(self):
"""finish processing.
add entries for sequences who only appear in the sbjct field.
"""
if not self.isSubset():
nids = self.mFasta.getContigSizes().keys()
nadded = 0
for nid in sorted(nids):
if nid not in self.mProfileLibrary:
self.applyMethod(AddaIO.NeighboursRecord(nid, []))
nadded += 1
self.mOutput += nadded
self.info("added %i profiles for sequences without neighbours" %
nadded)
self.mProfileLibrary.close()
AddaModuleRecord.finish(self)
config = AddaIO.ConfigParser()
config.read(os.path.expanduser(options.filename_config))
filename_graph = config.get("files", "output_graph", "adda.graph")
filename_index = config.get("files", "output_index", "adda.graph.index")
filename_fasta = config.get("files", "output_fasta", "adda")
fasta = IndexedFasta.IndexedFasta(filename_fasta)
index = cadda.IndexedNeighbours(filename_graph, filename_index)
config.set("files", "output_segments", "test.segments")
module = AddaSegment(
config=config,
fasta=fasta,
)
module.startUp()
args = map(int, args)
for nid in args:
neighbours = index.getNeighbours(nid)
module.applyMethod(AddaIO.NeighboursRecord(nid, neighbours))
module.finish
E.Stop()