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 SMART.Java.Python.structure.Transcript import Transcript
|
|
34 from commons.core.parsing.TranscriptListParser import TranscriptListParser
|
|
35
|
|
36
|
|
37 class GffParser(TranscriptListParser):
|
|
38 """A class that parses a GFF file and create a transcript list"""
|
|
39
|
|
40
|
|
41 def __init__(self, fileName, verbosity = 0):
|
|
42 super(GffParser, self).__init__(fileName, verbosity)
|
|
43
|
|
44
|
|
45 def __del__(self):
|
|
46 super(GffParser, self).__del__()
|
|
47
|
|
48
|
|
49 def getFileFormats():
|
|
50 return ["gff", "gff2", "gff3"]
|
|
51 getFileFormats = staticmethod(getFileFormats)
|
|
52
|
|
53
|
|
54 def skipFirstLines(self):
|
|
55 pass
|
|
56
|
|
57
|
|
58 def getInfos(self):
|
|
59 self.chromosomes = set()
|
|
60 self.nbTranscripts = 0
|
|
61 self.size = 0
|
|
62 self.reset()
|
|
63 if self.verbosity >= 10:
|
|
64 print "Getting information on %s." % (self.fileName)
|
|
65 self.reset()
|
|
66 for line in self.handle:
|
|
67 line = line.strip()
|
|
68 if line == "" or line[0] == "#":
|
|
69 continue
|
|
70 parts = line.split("\t")
|
|
71 if len(parts) != 9:
|
|
72 raise Exception("Error! Line '%s' has %d tab-separated fields instead of 9!" % (line, len(parts)))
|
|
73 self.chromosomes.add(parts[0])
|
|
74 if parts[8].find("Parent") == -1:
|
|
75 self.nbTranscripts += 1
|
|
76 else:
|
|
77 self.size += max(int(parts[3]), int(parts[4])) - min(int(parts[3]), int(parts[4])) + 1
|
|
78 if self.verbosity >= 10 and self.nbTranscripts % 100000 == 0:
|
|
79 sys.stdout.write(" %d transcripts read\r" % (self.nbTranscripts))
|
|
80 sys.stdout.flush()
|
|
81 self.reset()
|
|
82 if self.verbosity >= 10:
|
|
83 print " %d transcripts read" % (self.nbTranscripts)
|
|
84 print "Done."
|
|
85
|
|
86
|
|
87 def parseLine(self, line):
|
|
88 if not line or line[0] == "#":
|
|
89 return None
|
|
90 m = re.search(r"^\s*(\S+)\s+(\S+)\s+(\S+)\s+(\d+)\s+(\d+)\s+(\S+)\s+([+-.])\s+(\S+)\s+(\S.*)$", line)
|
|
91 if m == None:
|
|
92 raise Exception("\nLine %d '%s' does not have a GFF format\n" % (self.currentLineNb, line))
|
|
93 interval = Interval()
|
|
94 interval.setChromosome(m.group(1))
|
|
95 interval.setName("unnamed transcript")
|
|
96 interval.setStart(min(int(m.group(4)), int(m.group(5))))
|
|
97 interval.setEnd(max(int(m.group(4)), int(m.group(5))))
|
|
98 if m.group(7) == ".":
|
|
99 interval.setDirection("+")
|
|
100 else:
|
|
101 interval.setDirection(m.group(7))
|
|
102 interval.setTagValue("feature", m.group(3))
|
|
103 if m.group(6).isdigit():
|
|
104 interval.setTagValue("score", m.group(6))
|
|
105
|
|
106 remainings = m.group(9).split(";")
|
|
107 for remaining in remainings:
|
|
108 remaining = remaining.strip()
|
|
109 if remaining == "":
|
|
110 continue
|
|
111 posSpace = remaining.find(" ")
|
|
112 posEqual = remaining.find("=")
|
|
113 if posEqual != -1 and (posEqual < posSpace or posSpace == -1):
|
|
114 parts = remaining.split("=")
|
|
115 else:
|
|
116 parts = remaining.split()
|
|
117 field = parts[0].strip()
|
|
118 value = " ".join(parts[1:]).strip(" \"")
|
|
119 if field in ("Name", "name", "Sequence", "TE", "SAT"):
|
|
120 interval.setName(value)
|
|
121 else:
|
|
122 try:
|
|
123 intValue = int(value)
|
|
124 interval.setTagValue(field, intValue)
|
|
125 except ValueError:
|
|
126 interval.setTagValue(field, value)
|
|
127
|
|
128 self.currentTranscriptAddress = self.previousTranscriptAddress
|
|
129 if "Parent" in interval.getTagNames():
|
|
130 if self.currentTranscript == None:
|
|
131 raise Exception("GFF file does not start with a transcript! First line is '%s'." % (line.strip()))
|
|
132 if interval.getTagValue("Parent") != self.currentTranscript.getTagValue("ID"):
|
|
133 raise Exception("Exon '%s' is not right after its transcript in GFF file!" % (interval))
|
|
134 self.currentTranscript.addExon(interval)
|
|
135 if interval.name == None:
|
|
136 interval.name = self.currentTranscript.name
|
|
137 return None
|
|
138
|
|
139 transcript = self.currentTranscript
|
|
140 self.currentTranscript = Transcript()
|
|
141 self.currentTranscript.copy(interval)
|
|
142 self.previousTranscriptAddress = self.currentAddress
|
|
143
|
|
144 if transcript != None and transcript.name.startswith("unnamed"):
|
|
145 if "ID" in transcript.getTagNames():
|
|
146 transcript.name = transcript.getTagValue("ID")
|
|
147 else:
|
|
148 transcript.name = "unnamed transcript %s" % (self.currentLineNb)
|
|
149 return transcript
|