|
6
|
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 Clean a GFF file (as given by NCBI or TAIR) and outputs a GFF3 file.
|
|
|
33 """
|
|
|
34
|
|
|
35 from SMART.Java.Python.cleaning.TranscriptListCleaner import TranscriptListCleaner
|
|
|
36 from SMART.Java.Python.misc.Progress import Progress
|
|
|
37 from SMART.Java.Python.misc.UnlimitedProgress import UnlimitedProgress
|
|
|
38
|
|
|
39 count = {}
|
|
|
40
|
|
|
41 class ParsedLine(object):
|
|
|
42 def __init__(self, line, cpt):
|
|
|
43 self.line = line
|
|
|
44 self.cpt = cpt
|
|
|
45 self.parse()
|
|
|
46
|
|
|
47 def parse(self):
|
|
|
48 self.line = self.line.strip()
|
|
|
49 self.splittedLine = self.line.split(None, 8)
|
|
|
50 if len(self.splittedLine) < 9:
|
|
|
51 raise Exception("Line '%s' has less than 9 fields. Exiting..." % (self.line))
|
|
|
52 self.type = self.splittedLine[2]
|
|
|
53 self.parseOptions()
|
|
|
54 self.getId()
|
|
|
55 self.getParents()
|
|
|
56
|
|
|
57 def parseOptions(self):
|
|
|
58 self.parsedOptions = {}
|
|
|
59 for option in self.splittedLine[8].split(";"):
|
|
|
60 option = option.strip()
|
|
|
61 if option == "": continue
|
|
|
62 posSpace = option.find(" ")
|
|
|
63 posEqual = option.find("=")
|
|
|
64 if posEqual != -1 and (posEqual < posSpace or posSpace == -1):
|
|
|
65 key, value = option.split("=", 1)
|
|
|
66 elif posSpace != -1:
|
|
|
67 key, value = option.split(None, 1)
|
|
|
68 else:
|
|
|
69 key = "ID"
|
|
|
70 value = option
|
|
|
71 self.parsedOptions[key.strip()] = value.strip(" \"")
|
|
|
72
|
|
|
73 def getId(self):
|
|
|
74 for key in self.parsedOptions:
|
|
|
75 if key.lower() == "id":
|
|
|
76 self.id = self.parsedOptions[key]
|
|
|
77 return
|
|
|
78 if "Parent" in self.parsedOptions:
|
|
|
79 parent = self.parsedOptions["Parent"].split(",")[0]
|
|
|
80 if parent not in count:
|
|
|
81 count[parent] = {}
|
|
|
82 if self.type not in count[parent]:
|
|
|
83 count[parent][self.type] = 0
|
|
|
84 count[parent][self.type] += 1
|
|
|
85 self.id = "%s-%s-%d" % (parent, self.type, count[parent][self.type])
|
|
|
86 else:
|
|
|
87 self.id = "smart%d" % (self.cpt)
|
|
|
88 self.parsedOptions["ID"] = self.id
|
|
|
89
|
|
|
90 def getParents(self):
|
|
|
91 for key in self.parsedOptions:
|
|
|
92 if key.lower() in ("parent", "derives_from"):
|
|
|
93 self.parents = self.parsedOptions[key].split(",")
|
|
|
94 return
|
|
|
95 self.parents = None
|
|
|
96
|
|
|
97 def removeParent(self):
|
|
|
98 for key in self.parsedOptions.keys():
|
|
|
99 if key.lower() in ("parent", "derives_from"):
|
|
|
100 del self.parsedOptions[key]
|
|
|
101
|
|
|
102 def export(self):
|
|
|
103 self.splittedLine[8] = ";".join(["%s=%s" % (key, value) for key, value in self.parsedOptions.iteritems()])
|
|
|
104 return "%s\n" % ("\t".join(self.splittedLine))
|
|
|
105
|
|
|
106
|
|
|
107 class GffCleaner(TranscriptListCleaner):
|
|
|
108
|
|
|
109 def __init__(self, verbosity = 1):
|
|
|
110 super(GffCleaner, self).__init__(verbosity)
|
|
|
111 self.lines = {}
|
|
|
112 self.acceptedTypes = ["mRNA", "transcript", "exon"]
|
|
|
113 self.parents = []
|
|
|
114 self.children = {}
|
|
|
115
|
|
|
116 def getFileFormats():
|
|
|
117 return ["gff", "gff2", "gff3"]
|
|
|
118 getFileFormats = staticmethod(getFileFormats)
|
|
|
119
|
|
|
120 def setAcceptedTypes(self, types):
|
|
|
121 self.acceptedTypes = types
|
|
|
122
|
|
|
123 def parse(self):
|
|
|
124 progress = UnlimitedProgress(100000, "Reading input file", self.verbosity)
|
|
|
125 for cpt, line in enumerate(self.inputHandle):
|
|
56
|
126 if not line or (not line.strip()) or line[0] == "#": continue
|
|
6
|
127 if line[0] == ">": break
|
|
|
128 parsedLine = ParsedLine(line, cpt)
|
|
|
129 if self.acceptedTypes == None or parsedLine.type in self.acceptedTypes:
|
|
|
130 self.lines[parsedLine.id] = parsedLine
|
|
|
131 progress.inc()
|
|
|
132 progress.done()
|
|
|
133
|
|
|
134 def sort(self):
|
|
|
135 progress = Progress(len(self.lines.keys()), "Sorting file", self.verbosity)
|
|
|
136 for line in self.lines.values():
|
|
|
137 parentFound = False
|
|
|
138 if line.parents:
|
|
|
139 for parent in line.parents:
|
|
|
140 if parent in self.lines:
|
|
|
141 parentFound = True
|
|
|
142 if parent in self.children:
|
|
|
143 self.children[parent].append(line)
|
|
|
144 else:
|
|
|
145 self.children[parent] = [line]
|
|
|
146 if not parentFound:
|
|
|
147 line.removeParent()
|
|
|
148 self.parents.append(line)
|
|
|
149 progress.inc()
|
|
|
150 progress.done()
|
|
|
151
|
|
|
152 def write(self):
|
|
|
153 progress = Progress(len(self.parents), "Writing output file", self.verbosity)
|
|
|
154 for line in self.parents:
|
|
|
155 self.writeLine(line)
|
|
|
156 progress.inc()
|
|
|
157 progress.done()
|
|
|
158
|
|
|
159 def writeLine(self, line):
|
|
|
160 self.outputHandle.write(line.export())
|
|
|
161 if line.id in self.children:
|
|
|
162 for child in self.children[line.id]:
|
|
|
163 self.writeLine(child)
|
|
|
164
|
|
|
165 def _clean(self):
|
|
|
166 self.parse()
|
|
|
167 self.sort()
|
|
|
168 self.write()
|
