Mercurial > repos > yufei-luo > s_mart
comparison SMART/Java/Python/ncList/FindOverlapsWithSeveralIntervals.py @ 36:44d5973c188c
Uploaded
author | m-zytnicki |
---|---|
date | Tue, 30 Apr 2013 15:02:29 -0400 |
parents | 769e306b7933 |
children |
comparison
equal
deleted
inserted
replaced
35:d94018ca4ada | 36:44d5973c188c |
---|---|
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 | |
32 import os, struct, time | |
33 from optparse import OptionParser | |
34 from commons.core.parsing.ParserChooser import ParserChooser | |
35 from SMART.Java.Python.structure.Transcript import Transcript | |
36 from SMART.Java.Python.ncList.NCList import NCList | |
37 from SMART.Java.Python.ncList.NCListCursor import NCListCursor | |
38 from SMART.Java.Python.ncList.NCListFilePickle import NCListFilePickle, NCListFileUnpickle | |
39 from SMART.Java.Python.ncList.FileSorter import FileSorter | |
40 from SMART.Java.Python.misc.Progress import Progress | |
41 from SMART.Java.Python.misc.UnlimitedProgress import UnlimitedProgress | |
42 from SMART.Java.Python.ncList.NCListCursor import NCListCursor | |
43 from SMART.Java.Python.ncList.FindOverlapsWithOneInterval import FindOverlapsWithOneInterval | |
44 | |
45 REFERENCE = 0 | |
46 QUERY = 1 | |
47 TYPETOSTRING = {0: "reference", 1: "query"} | |
48 | |
49 class FindOverlapsWithSeveralIntervals(object): | |
50 | |
51 def __init__(self, verbosity = 1): | |
52 self._parsers = {} | |
53 self._outputFileName = "outputOverlaps.gff3" | |
54 self._iWriter = None | |
55 self._nbLines = {REFERENCE: 0, QUERY: 0} | |
56 self._verbosity = verbosity | |
57 self._ncLists = {} | |
58 self._sortedRefFileNames = None | |
59 self._transQueryFileName = None | |
60 self._cursors = {} | |
61 self._iFowoi = FindOverlapsWithOneInterval(self._verbosity) | |
62 | |
63 def __del__(self): | |
64 self.close() | |
65 for fileName in (self._sortedRefFileNames, self._transQueryFileName): | |
66 if os.path.exists(fileName): | |
67 os.remove(fileName) | |
68 | |
69 def close(self): | |
70 self._iFowoi.close() | |
71 | |
72 def setRefFileName(self, fileName, format): | |
73 self.setFileName(fileName, format, REFERENCE) | |
74 self._sortedRefFileNames = "%s_ref_sorted.pkl" % (os.path.splitext(fileName)[0]) | |
75 | |
76 def setQueryFileName(self, fileName, format): | |
77 self.setFileName(fileName, format, QUERY) | |
78 self._transQueryFileName = "%s_query_trans.pkl" % (os.path.splitext(fileName)[0]) | |
79 | |
80 def setFileName(self, fileName, format, type): | |
81 chooser = ParserChooser(self._verbosity) | |
82 chooser.findFormat(format) | |
83 self._parsers[type] = chooser.getParser(fileName) | |
84 | |
85 def setOutputFileName(self, outputFileName): | |
86 self._iFowoi.setOutputFileName(outputFileName) | |
87 | |
88 def _sortRefFile(self): | |
89 fs = FileSorter(self._parsers[REFERENCE], self._verbosity-4) | |
90 fs.perChromosome(True) | |
91 fs.setOutputFileName(self._sortedRefFileNames) | |
92 fs.sort() | |
93 self._nbLines[REFERENCE] = fs.getNbElements() | |
94 self._nbRefLinesPerChromosome = fs.getNbElementsPerChromosome() | |
95 self._splittedFileNames = fs.getOutputFileNames() | |
96 | |
97 def _translateQueryFile(self): | |
98 pickler = NCListFilePickle(self._transQueryFileName, self._verbosity) | |
99 progress = UnlimitedProgress(1000, "Translating query data", self._verbosity-4) | |
100 cpt = 0 | |
101 for queryTranscript in self._parsers[QUERY].getIterator(): | |
102 pickler.addTranscript(queryTranscript) | |
103 progress.inc() | |
104 cpt += 1 | |
105 progress.done() | |
106 self._nbLines[QUERY] = cpt | |
107 self._parsers[QUERY] = NCListFileUnpickle(self._transQueryFileName, self._verbosity) | |
108 | |
109 def prepareIntermediateFiles(self): | |
110 self._sortRefFile() | |
111 self._translateQueryFile() | |
112 | |
113 def createNCLists(self): | |
114 self._ncLists = {} | |
115 self._indices = {} | |
116 self._cursors = {} | |
117 for chromosome, fileName in self._splittedFileNames.iteritems(): | |
118 if self._verbosity > 3: | |
119 print " chromosome %s" % (chromosome) | |
120 ncList = NCList(self._verbosity) | |
121 ncList.createIndex(True) | |
122 ncList.setChromosome(chromosome) | |
123 ncList.setFileName(fileName) | |
124 ncList.setNbElements(self._nbRefLinesPerChromosome[chromosome]) | |
125 ncList.buildLists() | |
126 self._ncLists[chromosome] = ncList | |
127 cursor = NCListCursor(None, ncList, 0, self._verbosity) | |
128 self._cursors[chromosome] = cursor | |
129 self._indices[chromosome] = ncList.getIndex() | |
130 endTime = time.time() | |
131 | |
132 def compare(self): | |
133 progress = Progress(self._nbLines[QUERY], "Comparing data", self._verbosity-3) | |
134 startTime = time.time() | |
135 for cpt, queryTranscript in enumerate(self._parsers[QUERY].getIterator()): | |
136 chromosome = queryTranscript.getChromosome() | |
137 if chromosome not in self._ncLists: | |
138 continue | |
139 self._iFowoi.setNCList(self._ncLists[chromosome], self._indices[chromosome]) | |
140 self._iFowoi.setTranscript(queryTranscript) | |
141 self._iFowoi.compare() | |
142 self._iFowoi.dumpWriter() | |
143 progress.inc() | |
144 progress.done() | |
145 endTime = time.time() | |
146 self._timeSpent = endTime - startTime | |
147 | |
148 def run(self): | |
149 startTime = time.time() | |
150 if self._verbosity > 2: | |
151 print "Creating NC-list..." | |
152 self.prepareIntermediateFiles() | |
153 self.createNCLists() | |
154 endTime = time.time() | |
155 if self._verbosity > 2: | |
156 print " ...done (%.2gs)" % (endTime - startTime) | |
157 self.compare() | |
158 self.close() | |
159 if self._verbosity > 0: | |
160 print "# queries: %d" % (self._nbLines[QUERY]) | |
161 print "# refs: %d" % (self._nbLines[REFERENCE]) | |
162 print "# written: %d (%d overlaps)" % (self._iFowoi._nbWritten, self._iFowoi._nbOverlaps) | |
163 print "time: %.2gs" % (self._timeSpent) | |
164 | |
165 | |
166 if __name__ == "__main__": | |
167 description = "FindOverlaps With Several Intervals v1.0.0: Finds overlaps with several query intervals. [Category: Data comparison]" | |
168 | |
169 parser = OptionParser(description = description) | |
170 parser.add_option("-i", "--query", dest="inputQueryFileName", action="store", type="string", help="Query input file [compulsory] [format: file in transcript format given by -f]") | |
171 parser.add_option("-f", "--queryFormat", dest="queryFormat", action="store", type="string", help="format of previous file [compulsory] [format: transcript file format]") | |
172 parser.add_option("-j", "--ref", dest="inputRefFileName", action="store", type="string", help="Reference input file [compulsory] [format: file in transcript format given by -g]") | |
173 parser.add_option("-g", "--refFormat", dest="refFormat", action="store", type="string", help="format of previous file [compulsory] [format: transcript file format]") | |
174 parser.add_option("-o", "--output", dest="outputFileName", action="store", type="string", help="Output file [compulsory] [format: output file in GFF3 format]") | |
175 parser.add_option("-v", "--verbosity", dest="verbosity", action="store", default=1, type="int", help="Trace level [format: int] [default: 1]") | |
176 (options, args) = parser.parse_args() | |
177 | |
178 iFWSI = FindOverlapsWithSeveralIntervals(options.verbosity) | |
179 iFWSI.setRefFileName(options.inputRefFileName, options.refFormat) | |
180 iFWSI.setQueryFileName(options.inputQueryFileName, options.queryFormat) | |
181 iFWSI.setOutputFileName(options.outputFileName) | |
182 iFWSI.run() |