6
|
1 #
|
|
2 # Copyright INRA-URGI 2009-2010
|
|
3 #
|
|
4 # This software is governed by the CeCILL license under French law and
|
|
5 # abiding by the rules of distribution of free software. You can use,
|
|
6 # modify and/ or redistribute the software under the terms of the CeCILL
|
|
7 # license as circulated by CEA, CNRS and INRIA at the following URL
|
|
8 # "http://www.cecill.info".
|
|
9 #
|
|
10 # As a counterpart to the access to the source code and rights to copy,
|
|
11 # modify and redistribute granted by the license, users are provided only
|
|
12 # with a limited warranty and the software's author, the holder of the
|
|
13 # economic rights, and the successive licensors have only limited
|
|
14 # liability.
|
|
15 #
|
|
16 # In this respect, the user's attention is drawn to the risks associated
|
|
17 # with loading, using, modifying and/or developing or reproducing the
|
|
18 # software by the user in light of its specific status of free software,
|
|
19 # that may mean that it is complicated to manipulate, and that also
|
|
20 # therefore means that it is reserved for developers and experienced
|
|
21 # professionals having in-depth computer knowledge. Users are therefore
|
|
22 # encouraged to load and test the software's suitability as regards their
|
|
23 # requirements in conditions enabling the security of their systems and/or
|
|
24 # data to be ensured and, more generally, to use and operate it in the
|
|
25 # same conditions as regards security.
|
|
26 #
|
|
27 # The fact that you are presently reading this means that you have had
|
|
28 # knowledge of the CeCILL license and that you accept its terms.
|
|
29 #
|
|
30 import re
|
|
31 import sys
|
|
32 from SMART.Java.Python.structure.Interval import Interval
|
|
33 from commons.core.parsing.TranscriptListParser import TranscriptListParser
|
|
34 from SMART.Java.Python.structure.Transcript import Transcript
|
|
35
|
|
36
|
|
37 class BedParser(TranscriptListParser):
|
|
38 """A class that parses a BED file and create a transcript list"""
|
|
39
|
|
40
|
|
41 def __init__(self, fileName, verbosity = 0):
|
|
42 self.title = None
|
|
43 TranscriptListParser.__init__(self, fileName, verbosity)
|
|
44
|
|
45
|
|
46 # def __del__(self):
|
|
47 # super(BedParser, self).__del__()
|
|
48
|
|
49
|
|
50 def getFileFormats():
|
|
51 return ["bed"]
|
|
52 getFileFormats = staticmethod(getFileFormats)
|
|
53
|
|
54
|
|
55 def skipFirstLines(self):
|
|
56 mark = self.handle.tell()
|
|
57 line = self.handle.readline()
|
|
58 line = line.strip()
|
|
59 m = re.search(r"^\s*track\s+name\s*=\s*(\S+)\s+", line)
|
|
60 if m != None:
|
|
61 self.title = m.group(1)
|
|
62 self.currentLineNb += 1
|
|
63 else:
|
|
64 self.handle.seek(mark)
|
|
65 return
|
|
66
|
|
67
|
|
68
|
|
69
|
|
70 def parseLine(self, line):
|
|
71 m = re.search(r"^\s*(\S+)\t+(\d+)\t+(\d+)\s*$", line)
|
|
72 if m != None:
|
|
73 transcript = Transcript()
|
|
74 transcript.setChromosome(m.group(1))
|
|
75 transcript.setStart(min(int(m.group(2)), int(m.group(3))-1))
|
|
76 transcript.setEnd(max(int(m.group(2)), int(m.group(3))-1))
|
|
77 transcript.setName("Unnamed")
|
|
78 transcript.setDirection(1)
|
|
79 return transcript
|
|
80
|
|
81 m = re.search(r"^\s*(\S+)\t+(\d+)\t+(\d+)\t+([^\t]+)\s*$", line)
|
|
82 if m != None:
|
|
83 transcript = Transcript()
|
|
84 transcript.setChromosome(m.group(1))
|
|
85 transcript.setStart(min(int(m.group(2)), int(m.group(3))-1))
|
|
86 transcript.setEnd(max(int(m.group(2)), int(m.group(3))-1))
|
|
87 transcript.setName(m.group(4))
|
|
88 transcript.setDirection(1)
|
|
89 return transcript
|
|
90
|
|
91 m = re.search(r"^\s*(\S+)\t+(\d+)\t+(\d+)\t+([^\t]+)\t+\d+\.?\d*\s*$", line)
|
|
92 if m != None:
|
|
93 transcript = Transcript()
|
|
94 transcript.setChromosome(m.group(1))
|
|
95 transcript.setStart(min(int(m.group(2)), int(m.group(3))-1))
|
|
96 transcript.setEnd(max(int(m.group(2)), int(m.group(3))-1))
|
|
97 transcript.setName(m.group(4))
|
|
98 transcript.setDirection(1)
|
|
99 return transcript
|
|
100
|
|
101 m = re.search(r"^\s*(\S+)\t+(\d+)\t+(\d+)\t+([^\t]+)\t+\d+\t+([+-])\t+\d+\t+\d+\t+0\t+(\d+)\t+(\S+)\t+(\S+)\s*$", line)
|
|
102 if m == None:
|
|
103 raise Exception("\nLine %d '%s' does not has a BED format." % (self.currentLineNb, line))
|
|
104 transcript = Transcript()
|
|
105 transcript.setChromosome(m.group(1))
|
|
106 transcript.setStart(min(int(m.group(2)), int(m.group(3))-1))
|
|
107 transcript.setEnd(max(int(m.group(2)), int(m.group(3))-1))
|
|
108 transcript.setName(m.group(4))
|
|
109 transcript.setDirection(m.group(5))
|
|
110 nbExons = int(m.group(6))
|
|
111 sizes = m.group(7).split(",")
|
|
112 starts = m.group(8).split(",")
|
|
113
|
|
114 # check for comment in name
|
|
115 m = re.search(r"^([^\(]*)\((\S+)\)$", transcript.getName())
|
|
116 if m != None:
|
|
117 transcript.setName(m.group(1))
|
|
118 transcript.setTagValues(m.group(2), ";", "=")
|
|
119
|
|
120 # check for nb occurrences in name
|
|
121 m = re.search(r"(.*)-(\d+)$", transcript.getName())
|
|
122 if m != None:
|
|
123 transcript.setName(m.group(1))
|
|
124 transcript.setOccurrence(int(m.group(2)))
|
|
125
|
|
126 for i in range(nbExons):
|
|
127 exon = Interval(transcript)
|
|
128 exon.setStart(int(starts[i])+transcript.getStart())
|
|
129 exon.setEnd(transcript.getStart()+int(starts[i])+int(sizes[i])-1)
|
|
130 exon.setSize(int(sizes[i]))
|
|
131 transcript.addExon(exon)
|
|
132
|
|
133 if transcript.exons[0].getStart() != transcript.getStart():
|
|
134 sys.exit("There is something wrong with the start of transcript line '%s': transcript starts at %d whereas first exon starts at %d" % (line.strip(), transcript.start, transcript.exons[0].start))
|
|
135 if transcript.exons[-1].getEnd() != transcript.getEnd():
|
|
136 sys.exit("There is something wrong with the end of transcript line '%s': transcript ends at %d whereas last exon ends at %d" % (line.strip(), transcript.end, transcript.exons[-1].end))
|
|
137
|
|
138 return transcript
|
|
139
|