|
36
|
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 CompareOverlappingSmallRef(object):
|
|
|
63
|
|
|
64 def __init__(self, verbosity):
|
|
|
65 self.verbosity = verbosity
|
|
|
66 self.tableNames = {}
|
|
|
67 self.nbQueries = 0
|
|
|
68 self.nbRefs = 0
|
|
|
69 self.nbWritten = 0
|
|
|
70 self.nbOverlaps = 0
|
|
|
71 self.invert = False
|
|
|
72 self.antisense = False
|
|
|
73 self.collinear = False
|
|
|
74 self.distance = None
|
|
|
75 self.pcOverlapQuery = False
|
|
|
76 self.pcOverlapRef = False
|
|
55
|
77 self.minOverlap = False
|
|
36
|
78 self.bins = {}
|
|
|
79 self.notOverlapping = False
|
|
|
80
|
|
|
81 def setReferenceFile(self, fileName, format):
|
|
|
82 chooser = ParserChooser(self.verbosity)
|
|
|
83 chooser.findFormat(format)
|
|
|
84 self.refParser = chooser.getParser(fileName)
|
|
|
85
|
|
|
86 def setQueryFile(self, fileName, format):
|
|
|
87 chooser = ParserChooser(self.verbosity)
|
|
|
88 chooser.findFormat(format)
|
|
|
89 self.queryParser = chooser.getParser(fileName)
|
|
|
90
|
|
|
91 def setOutputFile(self, fileName):
|
|
|
92 self.writer = TranscriptWriter(fileName, "gff3", self.verbosity)
|
|
|
93
|
|
|
94 def setDistance(self, distance):
|
|
|
95 self.distance = distance
|
|
|
96
|
|
|
97 def setCollinear(self, boolean):
|
|
|
98 self.collinear = boolean
|
|
|
99
|
|
|
100 def setAntisense(self, boolean):
|
|
|
101 self.antisense = boolean
|
|
|
102
|
|
|
103 def setInvert(self, boolean):
|
|
|
104 self.invert = boolean
|
|
|
105
|
|
|
106 def setMinPercentOverlap(self, pcOverlapQuery, pcOverlapRef):
|
|
|
107 self.pcOverlapQuery = pcOverlapQuery
|
|
|
108 self.pcOverlapRef = pcOverlapRef
|
|
|
109
|
|
55
|
110 def setMinOverlap(self, minOverlap):
|
|
|
111 self.minOverlap = minOverlap
|
|
|
112
|
|
36
|
113 def includeNotOverlapping(self, boolean):
|
|
|
114 self.notOverlapping = boolean
|
|
|
115
|
|
|
116 def loadRef(self):
|
|
|
117 progress = UnlimitedProgress(10000, "Reading references", self.verbosity)
|
|
|
118 for transcript in self.refParser.getIterator():
|
|
|
119 if transcript.__class__.__name__ == "Mapping":
|
|
|
120 transcript = transcript.getTranscript()
|
|
|
121 transcript = self._alterTranscript(transcript, REFERENCE)
|
|
|
122 chromosome = transcript.getChromosome()
|
|
|
123 bin = getBin(transcript.getStart(), transcript.getEnd())
|
|
|
124 if chromosome not in self.bins:
|
|
|
125 self.bins[chromosome] = {}
|
|
|
126 if bin not in self.bins[chromosome]:
|
|
|
127 self.bins[chromosome][bin] = []
|
|
|
128 self.bins[chromosome][bin].append(transcript)
|
|
|
129 self.nbRefs += 1
|
|
|
130 progress.inc()
|
|
|
131 progress.done()
|
|
|
132
|
|
|
133 def _alterTranscript(self, transcript, type):
|
|
|
134 if type == REFERENCE:
|
|
|
135 if self.distance != None:
|
|
|
136 transcript.extendExons(self.distance)
|
|
|
137 return transcript
|
|
|
138
|
|
|
139 def _compareTwoTranscripts(self, queryTranscript, refTranscript):
|
|
|
140 if not queryTranscript.overlapWithExon(refTranscript):
|
|
|
141 return False
|
|
|
142 if self.collinear and queryTranscript.getDirection() != refTranscript.getDirection():
|
|
|
143 return False
|
|
|
144 if self.antisense and queryTranscript.getDirection() == refTranscript.getDirection():
|
|
|
145 return False
|
|
55
|
146 if self.minOverlap and not queryTranscript.overlapWithExon(refTranscript, self.minOverlap):
|
|
|
147 return False
|
|
36
|
148 querySize = queryTranscript.getSize()
|
|
|
149 if self.pcOverlapQuery and not queryTranscript.overlapWithExon(refTranscript, int(querySize * self.pcOverlapQuery / 100.0)):
|
|
|
150 return False
|
|
|
151 refSize = refTranscript.getSize()
|
|
|
152 if self.pcOverlapRef and not queryTranscript.overlapWithExon(refTranscript, int(refSize * self.pcOverlapRef / 100.0)):
|
|
|
153 return False
|
|
|
154 return True
|
|
|
155
|
|
|
156 def _compareTranscript(self, queryTranscript):
|
|
|
157 queryChromosome = queryTranscript.getChromosome()
|
|
|
158 if queryChromosome not in self.bins:
|
|
|
159 return []
|
|
|
160 queryStart = queryTranscript.getStart()
|
|
|
161 queryEnd = queryTranscript.getEnd()
|
|
|
162 bins = getOverlappingBins(queryStart, queryEnd)
|
|
|
163 overlaps = {}
|
|
|
164 for binRange in bins:
|
|
|
165 for bin in range(binRange[0], binRange[1]+1):
|
|
|
166 if bin not in self.bins[queryChromosome]:
|
|
|
167 continue
|
|
|
168 for refTranscript in self.bins[queryChromosome][bin]:
|
|
|
169 if self._compareTwoTranscripts(queryTranscript, refTranscript):
|
|
|
170 nbElements = int(float(refTranscript.getTagValue("nbElements"))) if "nbElements" in refTranscript.getTagNames() else 1
|
|
|
171 overlaps[refTranscript.getName()] = int(float(refTranscript.getTagValue("nbElements"))) if "nbElements" in refTranscript.getTagNames() else 1
|
|
|
172 self.nbOverlaps += nbElements
|
|
|
173 return overlaps
|
|
|
174
|
|
|
175 def _updateTranscript(self, queryTranscript, overlaps):
|
|
|
176 if overlaps:
|
|
46
|
177 queryTranscript.setTagValue("nbOverlaps", sum(overlaps.values()))
|
|
36
|
178 queryTranscript.setTagValue("overlapsWith", "--".join(overlaps.keys())[:100])
|
|
46
|
179 else:
|
|
|
180 queryTranscript.setTagValue("nbOverlaps", "0")
|
|
36
|
181 return queryTranscript
|
|
|
182
|
|
|
183 def compare(self):
|
|
|
184 progress = UnlimitedProgress(10000, "Comparing queries", self.verbosity)
|
|
|
185 for queryTranscript in self.queryParser.getIterator():
|
|
|
186 if queryTranscript.__class__.__name__ == "Mapping":
|
|
|
187 queryTranscript = queryTranscript.getTranscript()
|
|
|
188 progress.inc()
|
|
|
189 self.nbQueries += 1
|
|
|
190 overlaps = self._compareTranscript(queryTranscript)
|
|
|
191 if self.notOverlapping or (overlaps and not self.invert) or (not overlaps and self.invert):
|
|
|
192 if not self.invert:
|
|
|
193 queryTranscript = self._updateTranscript(queryTranscript, overlaps)
|
|
|
194 self.writer.addTranscript(queryTranscript)
|
|
|
195 self.nbWritten += 1
|
|
|
196 progress.done()
|
|
|
197 self.writer.close()
|
|
|
198
|
|
|
199 def displayResults(self):
|
|
|
200 if self.verbosity > 0:
|
|
|
201 print "# queries: %d" % (self.nbQueries)
|
|
|
202 print "# refs: %d" % (self.nbRefs)
|
|
|
203 print "# written: %d (%d overlaps)" % (self.nbWritten, self.nbOverlaps)
|
|
|
204
|
|
|
205 def run(self):
|
|
|
206 self.loadRef()
|
|
|
207 self.compare()
|
|
|
208 self.displayResults()
|
|
|
209
|
|
|
210 if __name__ == "__main__":
|
|
|
211
|
|
|
212 description = "Compare Overlapping Small Reference v1.0.1: Provide the queries that overlap with a reference, when the reference is small. [Category: Data Comparison]"
|
|
|
213
|
|
|
214 parser = OptionParser(description = description)
|
|
|
215 parser.add_option("-i", "--input1", dest="inputFileName1", action="store", type="string", help="query input file [compulsory] [format: file in transcript format given by -f]")
|
|
|
216 parser.add_option("-f", "--format1", dest="format1", action="store", type="string", help="format of previous file [compulsory] [format: transcript file format]")
|
|
|
217 parser.add_option("-j", "--input2", dest="inputFileName2", action="store", type="string", help="reference input file [compulsory] [format: file in transcript format given by -g]")
|
|
|
218 parser.add_option("-g", "--format2", dest="format2", action="store", type="string", help="format of previous file [compulsory] [format: transcript file format]")
|
|
|
219 parser.add_option("-o", "--output", dest="outputFileName", action="store", type="string", help="output file [format: output file in GFF3 format]")
|
|
46
|
220 parser.add_option("-O", "--notOverlapping", dest="notOverlapping", action="store_true", default=False, help="also output not overlapping data [format: bool] [default: false]")
|
|
|
221 parser.add_option("-d", "--distance", dest="distance", action="store", default=0, type="int", help="accept some distance between query and reference [format: int]")
|
|
|
222 parser.add_option("-c", "--collinear", dest="collinear", action="store_true", default=False, help="provide collinear features [format: bool] [default: false]")
|
|
|
223 parser.add_option("-a", "--antisense", dest="antisense", action="store_true", default=False, help="provide antisense features [format: bool] [default: false]")
|
|
|
224 parser.add_option("-p", "--pcOverlapQuery", dest="pcOverlapQuery", action="store", default=False, type="int", help="min. % overlap of the query [format: bool] [default: false]")
|
|
|
225 parser.add_option("-P", "--pcOverlapRef", dest="pcOverlapRef", action="store", default=False, type="int", help="min. % overlap of the reference [format: bool] [default: false]")
|
|
55
|
226 parser.add_option("-m", "--minOverlap", dest="minOverlap", action="store", default=False, type="int", help="min. # overlap between query and reference [format: bool] [default: false]")
|
|
46
|
227 parser.add_option("-x", "--exclude", dest="exclude", action="store_true", default=False, help="invert the match [format: bool] [default: false]")
|
|
|
228 parser.add_option("-v", "--verbosity", dest="verbosity", action="store", default=1, type="int", help="trace level [format: int]")
|
|
36
|
229 (options, args) = parser.parse_args()
|
|
|
230
|
|
|
231 cosr = CompareOverlappingSmallRef(options.verbosity)
|
|
|
232 cosr.setQueryFile(options.inputFileName1, options.format1)
|
|
|
233 cosr.setReferenceFile(options.inputFileName2, options.format2)
|
|
|
234 cosr.setOutputFile(options.outputFileName)
|
|
|
235 cosr.includeNotOverlapping(options.notOverlapping)
|
|
|
236 cosr.setDistance(options.distance)
|
|
|
237 cosr.setAntisense(options.antisense)
|
|
|
238 cosr.setInvert(options.exclude)
|
|
|
239 cosr.setMinPercentOverlap(options.pcOverlapQuery, options.pcOverlapRef)
|
|
55
|
240 cosr.setMinOverlap(options.minOverlap)
|
|
36
|
241 cosr.run()
|
|
|
242
|
