Mercurial > repos > yufei-luo > s_mart

comparison SMART/Java/Python/GetIntersection.py @ 18:94ab73e8a190

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author m-zytnicki
date Mon, 29 Apr 2013 03:20:15 -0400
parents
children
comparison
equal deleted inserted replaced
17:b0e8584489e6 18:94ab73e8a190
1 #! /usr/bin/env python
2 #
3 # Copyright INRA-URGI 2009-2011
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 from optparse import OptionParser
32 from commons.core.parsing.ParserChooser import ParserChooser
33 from commons.core.writer.TranscriptWriter import TranscriptWriter
34 from SMART.Java.Python.structure.Interval import Interval
35 from SMART.Java.Python.structure.Transcript import Transcript
36 from SMART.Java.Python.structure.Mapping import Mapping
37 from SMART.Java.Python.misc.Progress import Progress
38 from SMART.Java.Python.misc.UnlimitedProgress import UnlimitedProgress
39
40 MINBIN = 3
41 MAXBIN = 7
42 REFERENCE = 0
43 QUERY = 1
44
45 def getBin(start, end):
46 for i in range(MINBIN, MAXBIN + 1):
47 binLevel = 10 ** i
48 if int(start / binLevel) == int(end / binLevel):
49 return int(i * 10 ** (MAXBIN + 1) + int(start / binLevel))
50 return int((MAXBIN + 1) * 10 ** (MAXBIN + 1))
51
52 def getOverlappingBins(start, end):
53 array = []
54 bigBin = int((MAXBIN + 1) * 10 ** (MAXBIN + 1))
55 for i in range(MINBIN, MAXBIN + 1):
56 binLevel = 10 ** i
57 array.append((int(i * 10 ** (MAXBIN + 1) + int(start / binLevel)), int(i * 10 ** (MAXBIN + 1) + int(end / binLevel))))
58 array.append((bigBin, bigBin))
59 return array
60
61
62 class GetIntersection(object):
63
64 def __init__(self, verbosity):
65 self.verbosity = verbosity
66 self.nbQueries = 0
67 self.nbRefs = 0
68 self.nbWritten = 0
69 self.bins = {}
70
71 def setReferenceFile(self, fileName, format):
72 chooser = ParserChooser(self.verbosity)
73 chooser.findFormat(format)
74 self.refParser = chooser.getParser(fileName)
75
76 def setQueryFile(self, fileName, format):
77 chooser = ParserChooser(self.verbosity)
78 chooser.findFormat(format)
79 self.queryParser = chooser.getParser(fileName)
80
81 def setOutputFile(self, fileName):
82 self.writer = TranscriptWriter(fileName, "gff3", self.verbosity)
83
84 def loadRef(self):
85 progress = UnlimitedProgress(10000, "Reading references", self.verbosity)
86 for transcript in self.refParser.getIterator():
87 if transcript.__class__.__name__ == "Mapping":
88 transcript = transcript.getTranscript()
89 chromosome = transcript.getChromosome()
90 bin = getBin(transcript.getStart(), transcript.getEnd())
91 if chromosome not in self.bins:
92 self.bins[chromosome] = {}
93 if bin not in self.bins[chromosome]:
94 self.bins[chromosome][bin] = []
95 self.bins[chromosome][bin].append(transcript)
96 self.nbRefs += 1
97 progress.inc()
98 progress.done()
99
100 def _compareTranscript(self, queryTranscript):
101 queryChromosome = queryTranscript.getChromosome()
102 if queryChromosome not in self.bins:
103 return None
104 queryStart = queryTranscript.getStart()
105 queryEnd = queryTranscript.getEnd()
106 bins = getOverlappingBins(queryStart, queryEnd)
107 overlaps = []
108 for binRange in bins:
109 for bin in range(binRange[0], binRange[1]+1):
110 if bin not in self.bins[queryChromosome]:
111 continue
112 for refTranscript in self.bins[queryChromosome][bin]:
113 newTranscript = queryTranscript.getIntersection(refTranscript)
114 if newTranscript != None:
115 overlaps.append(newTranscript)
116 if not overlaps:
117 return None
118 newTranscript = overlaps[0]
119 for transcript in overlaps[1:]:
120 newTranscript.merge(transcript)
121 return newTranscript
122
123 def compare(self):
124 progress = UnlimitedProgress(10000, "Comparing queries", self.verbosity)
125 for queryTranscript in self.queryParser.getIterator():
126 if queryTranscript.__class__.__name__ == "Mapping":
127 queryTranscript = queryTranscript.getTranscript()
128 progress.inc()
129 self.nbQueries += 1
130 newTranscript = self._compareTranscript(queryTranscript)
131 if newTranscript != None:
132 self.writer.addTranscript(queryTranscript)
133 self.nbWritten += 1
134 progress.done()
135 self.writer.close()
136
137 def displayResults(self):
138 print "# queries: %d" % (self.nbQueries)
139 print "# refs: %d" % (self.nbRefs)
140 print "# written: %d" % (self.nbWritten)
141
142 def run(self):
143 self.loadRef()
144 self.compare()
145 self.displayResults()
146
147 if __name__ == "__main__":
148
149 description = "Get Intersection v1.0.0: Shrink the first data set so that all bases covered by the first data set is also covered by the second data set. [Category: Data Comparison]"
150
151 parser = OptionParser(description = description)
152 parser.add_option("-i", "--input1", dest="inputFileName1", action="store", type="string", help="query input file [compulsory] [format: file in transcript format given by -f]")
153 parser.add_option("-f", "--format1", dest="format1", action="store", type="string", help="format of previous file [compulsory] [format: transcript file format]")
154 parser.add_option("-j", "--input2", dest="inputFileName2", action="store", type="string", help="reference input file [compulsory] [format: file in transcript format given by -g]")
155 parser.add_option("-g", "--format2", dest="format2", action="store", type="string", help="format of previous file [compulsory] [format: transcript file format]")
156 parser.add_option("-o", "--output", dest="outputFileName", action="store", type="string", help="output file [format: output file in GFF3 format]")
157 parser.add_option("-v", "--verbosity", dest="verbosity", action="store", default=1, type="int", help="trace level [format: int]")
158 (options, args) = parser.parse_args()
159
160 gi = GetIntersection(options.verbosity)
161 gi.setQueryFile(options.inputFileName1, options.format1)
162 gi.setReferenceFile(options.inputFileName2, options.format2)
163 gi.setOutputFile(options.outputFileName)
164 gi.run()