| 6 | 1 #! /usr/bin/env python | 
|  | 2 # | 
|  | 3 # Copyright INRA-URGI 2009-2010 | 
|  | 4 # | 
|  | 5 # This software is governed by the CeCILL license under French law and | 
|  | 6 # abiding by the rules of distribution of free software. You can use, | 
|  | 7 # modify and/ or redistribute the software under the terms of the CeCILL | 
|  | 8 # license as circulated by CEA, CNRS and INRIA at the following URL | 
|  | 9 # "http://www.cecill.info". | 
|  | 10 # | 
|  | 11 # As a counterpart to the access to the source code and rights to copy, | 
|  | 12 # modify and redistribute granted by the license, users are provided only | 
|  | 13 # with a limited warranty and the software's author, the holder of the | 
|  | 14 # economic rights, and the successive licensors have only limited | 
|  | 15 # liability. | 
|  | 16 # | 
|  | 17 # In this respect, the user's attention is drawn to the risks associated | 
|  | 18 # with loading, using, modifying and/or developing or reproducing the | 
|  | 19 # software by the user in light of its specific status of free software, | 
|  | 20 # that may mean that it is complicated to manipulate, and that also | 
|  | 21 # therefore means that it is reserved for developers and experienced | 
|  | 22 # professionals having in-depth computer knowledge. Users are therefore | 
|  | 23 # encouraged to load and test the software's suitability as regards their | 
|  | 24 # requirements in conditions enabling the security of their systems and/or | 
|  | 25 # data to be ensured and, more generally, to use and operate it in the | 
|  | 26 # same conditions as regards security. | 
|  | 27 # | 
|  | 28 # The fact that you are presently reading this means that you have had | 
|  | 29 # knowledge of the CeCILL license and that you accept its terms. | 
|  | 30 # | 
|  | 31 """Find random regions in a genome""" | 
|  | 32 | 
|  | 33 import random, math | 
|  | 34 from optparse import OptionParser | 
|  | 35 from commons.core.parsing.FastaParser import * | 
|  | 36 from commons.core.writer.Gff3Writer import * | 
|  | 37 from commons.core.writer.MySqlTranscriptWriter import * | 
|  | 38 from SMART.Java.Python.misc.Progress import * | 
|  | 39 from SMART.Java.Python.structure.Transcript import Transcript | 
|  | 40 from SMART.Java.Python.structure.TranscriptContainer import TranscriptContainer | 
|  | 41 | 
|  | 42 repetitions = 100 | 
|  | 43 | 
|  | 44 | 
|  | 45 class RandomRegionsGenerator(object): | 
|  | 46 | 
|  | 47     def __init__(self, verbosity): | 
|  | 48         self.verbosity      = verbosity | 
|  | 49         self.strands        = False | 
|  | 50         self.distribution   = "uniform" | 
|  | 51         self.transcripts    = None | 
|  | 52         self.sequenceParser = None | 
|  | 53         random.seed() | 
|  | 54 | 
|  | 55 | 
|  | 56     def setInput(self, fileName): | 
|  | 57         self.sequenceParser = FastaParser(fileName, self.verbosity) | 
|  | 58 | 
|  | 59 | 
|  | 60     def setGenomeSize(self, size): | 
|  | 61         self.genomeSize = size | 
|  | 62 | 
|  | 63 | 
|  | 64     def setChromosomeName(self, name): | 
|  | 65         self.chromosomeName = name | 
|  | 66 | 
|  | 67 | 
|  | 68     def setAnnotation(self, fileName, format): | 
|  | 69         parser           = TranscriptContainer(fileName, format, self.verbosity) | 
|  | 70         self.transcripts = [] | 
|  | 71         for transcript in parser.getIterator(): | 
|  | 72             self.transcripts.append(transcript) | 
|  | 73         self.setNumber(len(self.transcripts)) | 
|  | 74         self.setSize(0) | 
|  | 75 | 
|  | 76 | 
|  | 77     def setOutputFile(self, fileName): | 
|  | 78         self.outputFileName = fileName | 
|  | 79 | 
|  | 80 | 
|  | 81     def setSize(self, size): | 
|  | 82         self.minSize = size | 
|  | 83         self.maxSize = size | 
|  | 84 | 
|  | 85 | 
|  | 86     def setMinSize(self, size): | 
|  | 87         self.minSize = size | 
|  | 88 | 
|  | 89 | 
|  | 90     def setMaxSize(self, size): | 
|  | 91         self.maxSize = size | 
|  | 92 | 
|  | 93 | 
|  | 94     def setNumber(self, number): | 
|  | 95         self.number = number | 
|  | 96 | 
|  | 97 | 
|  | 98     def setStrands(self, strands): | 
|  | 99         self.strands = strands | 
|  | 100 | 
|  | 101 | 
|  | 102     def setMaxDistribution(self, maxElements): | 
|  | 103         if maxElements == None: | 
|  | 104             return | 
|  | 105         self.maxElements = maxElements | 
|  | 106         self.distribution = "gaussian" | 
|  | 107 | 
|  | 108 | 
|  | 109     def setDeviationDistribution(self, deviation): | 
|  | 110         if deviation == None: | 
|  | 111             return | 
|  | 112         self.deviation = deviation | 
|  | 113         self.distribution = "gaussian" | 
|  | 114 | 
|  | 115 | 
|  | 116     def getSizes(self): | 
|  | 117         if self.sequenceParser == None: | 
|  | 118             self.chromosomes    = [self.chromosomeName] | 
|  | 119             self.sizes          = {self.chromosomeName: self.genomeSize} | 
|  | 120             self.cumulatedSize  = self.genomeSize | 
|  | 121             self.cumulatedSizes = {self.chromosomeName: self.genomeSize} | 
|  | 122             return | 
|  | 123         self.chromosomes    = self.sequenceParser.getRegions() | 
|  | 124         self.sizes          = {} | 
|  | 125         self.cumulatedSize  = 0 | 
|  | 126         self.cumulatedSizes = {} | 
|  | 127         for chromosome in self.chromosomes: | 
|  | 128             self.sizes[chromosome]          = self.sequenceParser.getSizeOfRegion(chromosome) | 
|  | 129             self.cumulatedSize             += self.sizes[chromosome] | 
|  | 130             self.cumulatedSizes[chromosome] = self.cumulatedSize | 
|  | 131 | 
|  | 132 | 
|  | 133     def findPosition(self, size = None): | 
|  | 134         if size == None: | 
|  | 135             size = random.randint(self.minSize, self.maxSize) | 
|  | 136         integer = random.randint(0, self.cumulatedSize) | 
|  | 137         for chromosome in self.chromosomes: | 
|  | 138             if self.cumulatedSizes[chromosome] > integer: | 
|  | 139                 break | 
|  | 140         start = random.randint(1, self.sizes[chromosome] - size) | 
|  | 141         return (chromosome, start, size) | 
|  | 142 | 
|  | 143 | 
|  | 144     def createTranscript(self, chromosome, start, size, strand, cpt): | 
|  | 145         transcript = Transcript() | 
|  | 146         transcript.setChromosome(chromosome) | 
|  | 147         transcript.setStart(start) | 
|  | 148         transcript.setEnd(start + size-1) | 
|  | 149         transcript.setDirection(strand) | 
|  | 150         transcript.setName("rand_%d" % (cpt)) | 
|  | 151         return transcript | 
|  | 152 | 
|  | 153 | 
|  | 154     def moveTranscript(self, chromosome, start, transcript): | 
|  | 155         while transcript.getEnd() + start - transcript.getStart() > self.cumulatedSizes[chromosome]: | 
|  | 156             chromosome, start, size = self.findPosition(transcript.getEnd() - transcript.getStart()) | 
|  | 157         transcript.setChromosome(chromosome) | 
|  | 158         oldStart, oldEnd = transcript.getStart(), transcript.getEnd() | 
|  | 159         if transcript.getNbExons() > 1: | 
|  | 160             for exon in transcript.getNbExons(): | 
|  | 161                 oldExonStart, oldExonEnd = exon.getStart(), exon.getEnd() | 
|  | 162                 exon.setStart(oldExonStart + start - oldStart) | 
|  | 163                 exon.setEnd(oldExonEnd + start - oldStart) | 
|  | 164         transcript.setStart(start) | 
|  | 165         transcript.setEnd(oldEnd + start - oldStart) | 
|  | 166         return [transcript] | 
|  | 167 | 
|  | 168 | 
|  | 169     def createUniformCluster(self, chromosome, start, size, strand, cpt): | 
|  | 170         transcript = self.createTranscript(chromosome, start, size, strand, cpt) | 
|  | 171         return [transcript] | 
|  | 172 | 
|  | 173 | 
|  | 174     def findNbTranscripts(self, cpt): | 
|  | 175         return min(int(round(math.exp(random.random() * math.log(self.maxElements)))), self.number - cpt + 1) | 
|  | 176 | 
|  | 177 | 
|  | 178     def getDev(self): | 
|  | 179         deviation = 0.0 | 
|  | 180         for j in range(repetitions): | 
|  | 181             deviation += random.randint(-self.deviation, self.deviation) | 
|  | 182         deviation /= repetitions | 
|  | 183         deviation  = int(round(deviation)) | 
|  | 184         return deviation | 
|  | 185 | 
|  | 186 | 
|  | 187     def createGaussianCluster(self, chromosome, start, size, strand, cpt): | 
|  | 188         transcripts   = [] | 
|  | 189         nbTranscripts = self.findNbTranscripts(cpt) | 
|  | 190         for i in range(nbTranscripts): | 
|  | 191             transcript = self.createTranscript(chromosome, start + self.getDev(), size + self.getDev(), strand, cpt + i) | 
|  | 192             transcripts.append(transcript) | 
|  | 193         return transcripts | 
|  | 194 | 
|  | 195 | 
|  | 196     def writeRegions(self): | 
|  | 197         writer     = Gff3Writer(self.outputFileName, self.verbosity) | 
|  | 198         outputFile = open(self.outputFileName, "w") | 
|  | 199         progress   = Progress(self.number, "Writing to %s" % (self.outputFileName), self.verbosity) | 
|  | 200         i          = 0 | 
|  | 201         while i < self.number: | 
|  | 202             chromosome, start, size = self.findPosition() | 
|  | 203             strand                  = random.choice([-1, 1]) if self.strands else 1 | 
|  | 204             if self.transcripts != None: | 
|  | 205                 transcripts = self.moveTranscript(chromosome, start, self.transcripts[i]) | 
|  | 206             elif self.distribution == "uniform": | 
|  | 207                 transcripts = self.createUniformCluster(chromosome, start, size, strand, i+1) | 
|  | 208             else: | 
|  | 209                 transcripts = self.createGaussianCluster(chromosome, start, size, strand, i+1) | 
|  | 210             for transcript in transcripts: | 
|  | 211                 writer.addTranscript(transcript) | 
|  | 212                 i += 1 | 
|  | 213                 progress.inc() | 
|  | 214         progress.done() | 
|  | 215         outputFile.close() | 
|  | 216         writer.write() | 
|  | 217         writer.close() | 
|  | 218 | 
|  | 219 | 
|  | 220     def run(self): | 
|  | 221         self.getSizes() | 
|  | 222         self.writeRegions() | 
|  | 223 | 
|  | 224 | 
|  | 225 if __name__ == "__main__": | 
|  | 226 | 
|  | 227     # parse command line | 
|  | 228     description = "Get Random Regions v1.0.2: Get some random coordinates on a genome. May use uniform or gaussian distribution (in gaussion distribution, # of element per cluster follows a power law). [Category: Other]" | 
|  | 229 | 
|  | 230     parser = OptionParser(description = description) | 
|  | 231     parser.add_option("-r", "--reference",     dest="reference",      action="store",      default=None,  type="string", help="file that contains the sequences [format: file in FASTA format]") | 
|  | 232     parser.add_option("-S", "--referenceSize", dest="referenceSize",  action="store",      default=None,  type="int",    help="size of the chromosome (when no reference is given) [format: int]") | 
|  | 233     parser.add_option("-c", "--chromosome",    dest="chromosome",     action="store",      default=None,  type="string", help="name of the chromosome (when no reference is given) [format: string]") | 
|  | 234     parser.add_option("-o", "--output",        dest="outputFileName", action="store",                     type="string", help="output file [compulsory] [format: output file in FASTA format]") | 
|  | 235     parser.add_option("-i", "--input",         dest="inputFileName",  action="store",      default=None,  type="string", help="optional file containing regions to shuffle [format: file in transcript format given by -f]") | 
|  | 236     parser.add_option("-f", "--format",        dest="format",         action="store",      default=None,  type="string", help="format of the previous file [format: transcript file format]") | 
|  | 237     parser.add_option("-s", "--size",          dest="size",           action="store",      default=None,  type="int",    help="size of the regions (if no region set is provided) [format: int]") | 
|  | 238     parser.add_option("-z", "--minSize",       dest="minSize",        action="store",      default=None,  type="int",    help="minimum size of the regions (if no region set nor a fixed size are provided) [format: int]") | 
|  | 239     parser.add_option("-Z", "--maxSize",       dest="maxSize",        action="store",      default=None,  type="int",    help="maximum size of the regions (if no region set nor a fixed size are provided) [format: int]") | 
|  | 240     parser.add_option("-n", "--number",        dest="number",         action="store",      default=None,  type="int",    help="number of regions (if no region set is provided) [format: int]") | 
|  | 241     parser.add_option("-t", "--strands",       dest="strands",        action="store_true", default=False,                help="use both strands (if no region set is provided) [format: boolean]") | 
|  | 242     parser.add_option("-m", "--max",           dest="max",            action="store",      default=None,  type="int",    help="max. # reads in a cluster (for Gaussian dist.) [format: int]") | 
|  | 243     parser.add_option("-d", "--deviation",     dest="deviation",      action="store",      default=None,  type="int",    help="deviation around the center of the cluster (for Gaussian dist.) [format: int]") | 
|  | 244     parser.add_option("-v", "--verbosity",     dest="verbosity",      action="store",      default=1,     type="int",    help="trace level [format: int]") | 
|  | 245     (options, args) = parser.parse_args() | 
|  | 246 | 
|  | 247     rrg = RandomRegionsGenerator(options.verbosity) | 
|  | 248     if options.reference == None: | 
|  | 249         rrg.setGenomeSize(options.referenceSize) | 
|  | 250         rrg.setChromosomeName(options.chromosome) | 
|  | 251     else: | 
|  | 252         rrg.setInput(options.reference) | 
|  | 253     rrg.setOutputFile(options.outputFileName) | 
|  | 254     if options.inputFileName == None: | 
|  | 255         if options.size != None: | 
|  | 256             rrg.setSize(options.size) | 
|  | 257         else: | 
|  | 258             rrg.setMinSize(options.minSize) | 
|  | 259             rrg.setMaxSize(options.maxSize) | 
|  | 260         rrg.setNumber(options.number) | 
|  | 261         rrg.setStrands(options.strands) | 
|  | 262     else: | 
|  | 263         rrg.setAnnotation(options.inputFileName, options.format) | 
|  | 264     rrg.setMaxDistribution(options.max) | 
|  | 265     rrg.setDeviationDistribution(options.deviation) | 
|  | 266     rrg.run() | 
|  | 267 |