view SMART/Java/Python/clusterize.py @ 71:d96f6c9a39e0 draft default tip

Removed pyc files.
author m-zytnicki
date Thu, 07 Apr 2016 09:25:18 -0400
parents f4de72c80eac
children
line wrap: on
line source

#! /usr/bin/env python
#
# Copyright INRA-URGI 2009-2010
# 
# This software is governed by the CeCILL license under French law and
# abiding by the rules of distribution of free software. You can use,
# modify and/ or redistribute the software under the terms of the CeCILL
# license as circulated by CEA, CNRS and INRIA at the following URL
# "http://www.cecill.info".
# 
# As a counterpart to the access to the source code and rights to copy,
# modify and redistribute granted by the license, users are provided only
# with a limited warranty and the software's author, the holder of the
# economic rights, and the successive licensors have only limited
# liability.
# 
# In this respect, the user's attention is drawn to the risks associated
# with loading, using, modifying and/or developing or reproducing the
# software by the user in light of its specific status of free software,
# that may mean that it is complicated to manipulate, and that also
# therefore means that it is reserved for developers and experienced
# professionals having in-depth computer knowledge. Users are therefore
# encouraged to load and test the software's suitability as regards their
# requirements in conditions enabling the security of their systems and/or
# data to be ensured and, more generally, to use and operate it in the
# same conditions as regards security.
# 
# The fact that you are presently reading this means that you have had
# knowledge of the CeCILL license and that you accept its terms.
#
from commons.core.writer.WriterChooser import WriterChooser
"""Clusterize a set of transcripts"""

import os, os.path, random
from optparse import OptionParser
from heapq import heappush, heappop
from commons.core.parsing.ParserChooser import ParserChooser
from commons.core.writer.Gff3Writer import Gff3Writer
from SMART.Java.Python.structure.Transcript import Transcript
from SMART.Java.Python.ncList.NCListFilePickle import NCListFileUnpickle
from SMART.Java.Python.ncList.FileSorter import FileSorter
from SMART.Java.Python.misc.Progress import Progress
from SMART.Java.Python.misc.UnlimitedProgress import UnlimitedProgress

class Clusterize(object):

	def __init__(self, verbosity):
		self.parsers           = {}
		self.sortedFileNames   = {}
		self.normalize         = False
		self.presorted         = False
		self.distance          = 1
		self.collinear         = False
		self.nbWritten         = 0
		self.nbMerges          = 0
		self.verbosity         = verbosity
		self.splittedFileNames = {}
		self.chromosomes       = set()

	def __del__(self):
		for fileName1 in self.splittedFileNames:
			for fileName2 in self.splittedFileNames[fileName1].values():
				os.remove(fileName2)

	def setInputFiles(self, fileNames, format):
		parserChooser = ParserChooser(self.verbosity)
		parserChooser.findFormat(format)
		for fileName in fileNames:
			self.parsers[fileName] = parserChooser.getParser(fileName)
			self.sortedFileNames[fileName] = "%s_sorted_%d.pkl" % (os.path.splitext(fileName)[0], random.randint(1, 100000))
			if "SMARTTMPPATH" in os.environ:
				self.sortedFileNames[fileName] = os.path.join(os.environ["SMARTTMPPATH"], os.path.basename(self.sortedFileNames[fileName]))

	def setOutputFileName(self, fileName, format="gff3", title="S-MART", feature="transcript", featurePart="exon"):
		writerChooser = WriterChooser()
		writerChooser.findFormat(format)
		self.writer = writerChooser.getWriter(fileName)
		self.writer.setTitle(title)
		self.writer.setFeature(feature)
		self.writer.setFeaturePart(featurePart)

	def setDistance(self, distance):
		self.distance = distance

	def setColinear(self, collinear):
		self.collinear = collinear

	def setNormalize(self, normalize):
		self.normalize = normalize
		
	def setPresorted(self, presorted):
		self.presorted = presorted

	def _sortFiles(self):
		if self.presorted:
			return
		for fileName, parser in self.parsers.iteritems():
			fs = FileSorter(parser, self.verbosity-4)
			fs.perChromosome(True)
			fs.setPresorted(self.presorted)
			fs.setOutputFileName(self.sortedFileNames[fileName])
			fs.sort()
			self.splittedFileNames[fileName] = fs.getOutputFileNames()
			self.chromosomes.update(self.splittedFileNames[fileName].keys())
		
	def _iterate(self):
		progress = UnlimitedProgress(10000, "Reading input file", self.verbosity)
		parsersSets = []
		self.nbElements = 0
		if self.chromosomes:
			for chromosome in self.chromosomes:
				parsersSets.append([NCListFileUnpickle(self.splittedFileNames[fileName][chromosome]) for fileName in self.splittedFileNames if chromosome in self.splittedFileNames[fileName]])
		else:
			parsersSets.append(self.parsers.values())
		for parsers in parsersSets:
			transcripts = []
			heap        = []
			for parser in parsers:
				iterator = parser.getIterator()
				for transcript in iterator:
					if transcript.__class__.__name__ == "Mapping":
						transcript = transcript.getTranscript()
					heappush(heap, (transcript.getChromosome(), transcript.getStart(), -transcript.getEnd(), transcript, iterator))
					break
			while heap:
				chromosome, start, end, newTranscript, iterator = heappop(heap)
				for transcript in iterator:
					if transcript.__class__.__name__ == "Mapping":
						transcript = transcript.getTranscript()
					heappush(heap, (transcript.getChromosome(), transcript.getStart(), -transcript.getEnd(), transcript, iterator))
					break
				newTranscripts = []
				if newTranscript.__class__.__name__ == "Mapping":
					newTranscript = newTranscript.getTranscript()
				for oldTranscript in transcripts:
					if self._checkOverlap(newTranscript, oldTranscript):
						self._merge(newTranscript, oldTranscript)
					elif self._checkPassed(newTranscript, oldTranscript):
						self._write(oldTranscript)
					else:
						newTranscripts.append(oldTranscript)
				newTranscripts.append(newTranscript)
				transcripts = newTranscripts
				self.nbElements += 1
				progress.inc()
			for transcript in transcripts:
				self._write(transcript)
		progress.done()

	def _merge(self, transcript1, transcript2):
		self.nbMerges += 1
		transcript2.setDirection(transcript1.getDirection())
		transcript1.merge(transcript2)

	def _write(self, transcript):
		self.nbWritten += 1
		self.writer.addTranscript(transcript)

	def _checkOverlap(self, transcript1, transcript2):
		if transcript1.getChromosome() != transcript2.getChromosome():
			return False
		if self.collinear and transcript1.getDirection() != transcript2.getDirection():
			return False
		if transcript1.getDistance(transcript2) > self.distance:
			return False
		return True

	def _checkPassed(self, transcript1, transcript2):
		return ((transcript1.getChromosome() != transcript2.getChromosome()) or (transcript1.getDistance(transcript2) > self.distance))

	def run(self):
		self._sortFiles()
		self._iterate()
		self.writer.close()
		if self.verbosity > 0:
			print "# input:   %d" % (self.nbElements)
			print "# written: %d (%d%% overlaps)" % (self.nbWritten, 0 if (self.nbElements == 0) else ((float(self.nbWritten) / self.nbElements) * 100))
			print "# merges:  %d" % (self.nbMerges)
		

if __name__ == "__main__":
	description = "Clusterize v1.0.3: clusterize the data which overlap. [Category: Merge]"

	parser = OptionParser(description = description)
	parser.add_option("-i", "--inputs",       dest="inputFileNames", action="store",				     type="string", help="input files (separated by commas) [compulsory] [format: string]")
	parser.add_option("-f", "--format",       dest="format",		 action="store",				     type="string", help="format of file [format: transcript file format]")
	parser.add_option("-o", "--output",       dest="outputFileName", action="store",				     type="string", help="output file [compulsory] [format: output file in transcript format given by -u]")
	parser.add_option("-u", "--outputFormat", dest="outputFormat",   action="store",      default="gff",		        type="string", help="output file format [format: transcript file format]")
	parser.add_option("-c", "--collinear",    dest="collinear",      action="store_true", default=False,				help="merge collinear transcripts only [format: bool] [default: false]")
	parser.add_option("-d", "--distance",     dest="distance",       action="store",      default=0,     type="int",    help="max. distance between two transcripts to be merged [format: int] [default: 0]")
	parser.add_option("-n", "--normalize",    dest="normalize",      action="store_true", default=False,				help="normalize the number of reads per cluster by the number of mappings per read [format: bool] [default: false]")
	parser.add_option("-s", "--sorted",       dest="sorted",		 action="store_true", default=False,				help="input is already sorted [format: bool] [default: false]")
	parser.add_option("-v", "--verbosity",    dest="verbosity",      action="store",      default=1,     type="int",    help="trace level [format: int] [default: 1]")
	(options, args) = parser.parse_args()

	c = Clusterize(options.verbosity)
	c.setInputFiles(options.inputFileNames.split(","), options.format)
	c.setOutputFileName(options.outputFileName, options.outputFormat)
	c.setColinear(options.collinear)
	c.setDistance(options.distance)
	c.setNormalize(options.normalize)
	c.setPresorted(options.sorted)
	c.run()