Mercurial > repos > yufei-luo > s_mart
view SMART/Java/Python/RestrictFromCoverage.py @ 20:c265f13066d8
Deleted selected files
author | m-zytnicki |
---|---|
date | Mon, 29 Apr 2013 03:24:43 -0400 |
parents | 769e306b7933 |
children |
line wrap: on
line source
#! /usr/bin/env python # # Copyright INRA-URGI 2009-2012 # # 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. # import os, struct, time, random from optparse import OptionParser 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.structure.Interval import Interval from SMART.Java.Python.ncList.NCList import NCList from SMART.Java.Python.ncList.NCListCursor import NCListCursor from SMART.Java.Python.ncList.NCListFilePickle import NCListFilePickle, 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 from SMART.Java.Python.misc import Utils try: import cPickle as pickle except: import pickle REFERENCE = 0 QUERY = 1 TYPES = (REFERENCE, QUERY) TYPETOSTRING = {0: "reference", 1: "query"} class RestrictFromCoverage(object): def __init__(self, verbosity = 1): self._verbosity = verbosity self._randomNumber = random.randint(0, 100000) self._nbWritten = 0 self._nbLines = dict([type, 0] for type in TYPES) self._splittedFileNames = dict([type, {}] for type in TYPES) self._nbElementsPerChromosome = dict([type, {}] for type in TYPES) self._nbElements = dict([type, 0] for type in TYPES) def __del__(self): pass def _close(self): self._writer.close() def setInputFileName(self, fileName, format, type): chooser = ParserChooser(self._verbosity) chooser.findFormat(format) parser = chooser.getParser(fileName) sortedFileName = "%s_%d_%d_sorted.pkl" % (os.path.splitext(fileName)[0], self._randomNumber, type) if self._verbosity > 2: print "Preparing %s file..." % (TYPETOSTRING[type]) startTime = time.time() fs = FileSorter(parser, self._verbosity-1) fs.perChromosome(True) fs.setOutputFileName(sortedFileName) fs.sort() self._nbLines[type] = fs.getNbElements() self._splittedFileNames[type] = fs.getOutputFileNames() self._nbElementsPerChromosome[type] = fs.getNbElementsPerChromosome() self._nbElements[type] = fs.getNbElements() endTime = time.time() if self._verbosity > 2: print " ...done (%ds)" % (endTime - startTime) def setOutputFileName(self, outputFileName): self._writer = Gff3Writer(outputFileName) def setPercent(self, minPercent, maxPercent): self._minPercent = minPercent self._maxPercent = maxPercent def setNbNucleotides(self, minNb, maxNb): self._minNucleotides = minNb self._maxNucleotides = maxNb def setOverlap(self, minOverlap, maxOverlap): self._minOverlap = minOverlap self._maxOverlap = maxOverlap def setStrands(self, boolean): self._twoStrands = boolean def _compareChromosome(self, chromosome): firstOverlap = 0 parser1 = NCListFileUnpickle(self._splittedFileNames[QUERY][chromosome], self._verbosity) parser2 = NCListFileUnpickle(self._splittedFileNames[REFERENCE][chromosome], self._verbosity) progress = Progress(self._nbElementsPerChromosome[QUERY][chromosome], "Analyzing %s" % (chromosome), self._verbosity) for transcript1 in parser1.getIterator(): firstOverlap = self._compareList(transcript1, parser2) parser2.setInitAddress(firstOverlap) progress.inc() progress.done() def _compareList(self, transcript1, parser2): values = [] for exon in transcript1.getExons(): values.append([0.0] * exon.getSize()) firstOverlap = None for transcript2 in parser2.getIterator(): address = parser2.getCurrentTranscriptAddress() nbElements = float(transcript2.getTagValue("nbElements")) if "nbElements" in transcript2.getTagNames() else 1.0 nbOccurrences = float(transcript2.getTagValue("nbOccurrences")) if "nbOccurrences" in transcript2.getTagNames() else 1.0 nbElements /= nbOccurrences if transcript2.getStart() > transcript1.getEnd(): if firstOverlap == None: firstOverlap = address if self._checkValues(values): self._printTranscript(transcript1) return firstOverlap elif transcript1.overlapWith(transcript2): if firstOverlap == None: firstOverlap = address values = self._compareTranscript(transcript1, transcript2, values, nbElements) if self._checkValues(values): self._printTranscript(transcript1) return firstOverlap def _compareTranscript(self, transcript1, transcript2, values, nbElements): if not transcript1.overlapWith(transcript2) or ((self._twoStrands) and transcript1.getDirection() != transcript2.getDirection()): return values for id1, exon1 in enumerate(transcript1.getExons()): for exon2 in transcript2.getExons(): values[id1] = map(sum, zip(values[id1], self._compareExon(exon1, exon2, nbElements))) return values def _compareExon(self, exon1, exon2, nbElements): array = [0.0] * exon1.getSize() if not exon1.overlapWith(exon2) or ((self._twoStrands) and exon1.getDirection() != exon2.getDirection()): return array for pos in range(max(exon1.getStart(), exon2.getStart()) - exon1.getStart(), min(exon1.getEnd(), exon2.getEnd()) - exon1.getStart()+1): array[pos] += nbElements return array def _filter(self, value): if self._minOverlap and self._maxOverlap: return self._minOverlap <= value <= self._maxOverlap if self._minOverlap: return self._minOverlap <= value if self._maxOverlap: return value <= self._maxOverlap return True def _checkValues(self, values): nbValues = sum(map(len, values)) nbPosValues = sum(map(len, [filter(self._filter, valuePart) for valuePart in values])) ratio = float(nbPosValues) / nbValues * 100 if self._minNucleotides and nbPosValues < self._minNucleotides: return False if self._maxNucleotides and nbPosValues > self._maxNucleotides: return False if self._minPercent and ratio < self._minPercent: return False if self._maxPercent and ratio > self._maxPercent: return False return True def _printTranscript(self, transcript): self._writer.addTranscript(transcript) self._nbWritten += 1 def run(self): for chromosome in sorted(self._splittedFileNames[QUERY].keys()): self._compareChromosome(chromosome) self._close() if self._verbosity > 0: print "# queries: %d" % (self._nbElements[QUERY]) print "# refs: %d" % (self._nbElements[REFERENCE]) print "# written: %d (%d%%)" % (self._nbWritten, 0 if self._nbElements[QUERY] == 0 else round(float(self._nbWritten) / self._nbElements[QUERY] * 100)) if __name__ == "__main__": description = "Restrict From Coverage v1.0.0: Select the elements from the first set which have a given coverage. [Category: Data Comparison]" parser = OptionParser(description = description) parser.add_option("-i", "--input1", dest="inputFileName1", action="store", type="string", help="input file 1 [compulsory] [format: file in transcript format given by -f]") parser.add_option("-f", "--format1", dest="format1", action="store", type="string", help="format of file 1 [compulsory] [format: transcript file format]") parser.add_option("-j", "--input2", dest="inputFileName2", action="store", type="string", help="input file 2 [compulsory] [format: file in transcript format given by -g]") parser.add_option("-g", "--format2", dest="format2", action="store", type="string", help="format of file 2 [compulsory] [format: transcript file format]") parser.add_option("-o", "--output", dest="output", action="store", default=None, type="string", help="output file [compulsory] [format: output file in GFF3 format]") parser.add_option("-n", "--minNucleotides", dest="minNucleotides", action="store", default=None, type="int", help="minimum number of nucleotides overlapping to declare an overlap [format: int]") parser.add_option("-N", "--maxNucleotides", dest="maxNucleotides", action="store", default=None, type="int", help="maximum number of nucleotides overlapping to declare an overlap [format: int]") parser.add_option("-p", "--minPercent", dest="minPercent", action="store", default=None, type="int", help="minimum percentage of nucleotides overlapping to declare an overlap [format: int]") parser.add_option("-P", "--maxPercent", dest="maxPercent", action="store", default=None, type="int", help="maximum percentage of nucleotides overlapping to declare an overlap [format: int]") parser.add_option("-e", "--minOverlap", dest="minOverlap", action="store", default=None, type="int", help="minimum number of elements from 2nd file to declare an overlap [format: int]") parser.add_option("-E", "--maxOverlap", dest="maxOverlap", action="store", default=None, type="int", help="maximum number of elements from 2nd file to declare an overlap [format: int]") parser.add_option("-s", "--strands", dest="strands", action="store_true", default=False, help="consider the two strands separately [format: bool] [default: false]") parser.add_option("-v", "--verbosity", dest="verbosity", action="store", default=1, type="int", help="trace level [format: int]") (options, args) = parser.parse_args() rfc = RestrictFromCoverage(options.verbosity) rfc.setInputFileName(options.inputFileName1, options.format1, QUERY) rfc.setInputFileName(options.inputFileName2, options.format2, REFERENCE) rfc.setOutputFileName(options.output) rfc.setNbNucleotides(options.minNucleotides, options.maxNucleotides) rfc.setPercent(options.minPercent, options.maxPercent) rfc.setOverlap(options.minOverlap, options.maxOverlap) rfc.setStrands(options.strands) rfc.run()