Debian Codegraph @

csb/bio/io/cs.py

raw
1 """
2 Simple NMR STAR chemical shift readers.
3 """
4
5 from csb.bio.nmr import ChemShiftInfo
6
7
8 class ChemShiftFormatError(ValueError):
9 pass
10
11
12 class ChemShiftReader(object):
13 """
14 Simple NMR STAR v2 chemical shift reader.
15
16 @note: This is not a full-fledged, semantic NMR STAR parser. It handles
17 only the chemical shift table.
18 """
19
20 FRAME = 'save_assigned_chemical_shifts'
21
22 RANK = '_Residue_seq_code'
23 RESIDUE = '_Residue_label'
24 ATOM = '_Atom_name'
25 ELEMENT = '_Atom_type'
26 SHIFT = '_Chem_shift_value'
27
28 @staticmethod
29 def create(frame=FRAME, version=2):
30 """
31 Parser factory: create a new parser, given a saveframe name
32 and format verison.
33
34 @param frame: name of the saveframe to read
35 @type frame: str
36 @param version: NMR STAR format version
37 @type version: int
38
39 @return: an instance of any L{ChemShiftReader} class
40 @rtype: L{ChemShiftReader}
41 """
42
43 if version == 3:
44 return ChemShift3Reader(frame=frame)
45 elif version == 2:
46 return ChemShiftReader(frame=frame)
47 else:
48 raise ValueError('Unknown NMR-STAR version')
49
50 @staticmethod
51 def guess(file, frame=FRAME):
52 """
53 Parser factory: try to guess the correct NMR STAR version from a given
54 file and create an appropriate parser.
55
56 @param file: NMR STAR path and file name
57 @type file: str
58 @param frame: name of the saveframe to read
59 @type frame: str
60
61 @return: an instance of any L{ChemShiftReader} class
62 @rtype: L{ChemShiftReader}
63
64 @raise ChemShiftFormatError: on failure to determine the NMR STAR version
65 """
66
67 with open(file) as cs:
68 content = cs.read()
69
70 if not content.strip():
71 return ChemShiftReader.create()
72 elif ChemShift3Reader.SHIFT3 in content:
73 return ChemShiftReader.create(frame, version=3)
74 elif ChemShiftReader.SHIFT in content:
75 return ChemShiftReader.create(frame, version=2)
76 else:
77 raise ChemShiftFormatError("Can't guess NMR-STAR version")
78
79 def __init__(self, frame=FRAME):
80 self._frame = frame
81
82 def read_file(self, filename):
83 """
84 Parse the specified file.
85
86 @param filename: file path and name
87 @type filename: str
88
89 @rtype: tuple of L{ChemShiftInfo}
90 """
91 with open(filename) as input:
92 return self.read_shifts(input.read())
93
94 def read_shifts(self, star_table):
95 """
96 Parse a given NMR STAR chemical shift table.
97
98 @param star_table: NMR STAR chemical shift table
99 @type star_table: str
100
101 @rtype: tuple of L{ChemShiftInfo}
102 @raise ChemShiftFormatError: on parse error
103 """
104
105 shifts = []
106
107 init = False
108 in_shifts = False
109 fields = []
110 lines = iter(star_table.splitlines())
111
112 if self._frame in star_table:
113 self._scroll(lines, self._frame)
114
115
116 for l in lines:
117 ls = l.strip()
118
119 if not in_shifts:
120
121 if ls == 'loop_':
122 assert in_shifts is False and not fields and init is False
123 init = True
124 continue
125
126 elif init and ls.startswith('_'):
127 assert in_shifts is False
128 fields.append(l.strip())
129 continue
130
131 elif init and not ls:
132 if len(fields) < 1:
133 raise ChemShiftFormatError("No fields found in the CS table")
134 in_shifts = True
135 continue
136
137 else:
138
139 if ls == 'stop_':
140 break
141
142 elif ls.startswith('#'):
143 continue
144
145 elif ls:
146 values = l.split()
147 if len(values) < len(fields):
148 raise ChemShiftFormatError("Insufficient number of values: {0}".format(l))
149 data = dict(zip(fields, values))
150
151 shifts.append(self._create_shift(data))
152
153 return tuple(shifts)
154
155 def _scroll(self, iterator, field):
156
157 for line in iterator:
158 if line.lstrip().startswith(field):
159 break
160
161 def _create_shift(self, data):
162
163 try:
164 position = int(data[ChemShiftReader.RANK])
165 residue = data[ChemShiftReader.RESIDUE]
166 name = data[ChemShiftReader.ATOM]
167 element = data[ChemShiftReader.ELEMENT]
168 shift = float(data[ChemShiftReader.SHIFT])
169
170 except KeyError as ke:
171 raise ChemShiftFormatError("Required field {0} not found".format(str(ke)))
172 except ValueError as ve:
173 raise ChemShiftFormatError("Can't parse value: {0}".format(str(ve)))
174
175 return ChemShiftInfo(position, residue, name, element, shift)
176
177
178 class ChemShift3Reader(ChemShiftReader):
179 """
180 Simple NMR STAR v3 chemical shift reader.
181
182 @note: This is not a full-fledged, semantic NMR STAR parser. It handles
183 only the chemical shift table.
184 """
185
186 RANK3 = '_Atom_chem_shift.Seq_ID'
187 RESIDUE3 = '_Atom_chem_shift.Comp_ID'
188 ATOM3 = '_Atom_chem_shift.Atom_ID'
189 ELEMENT3 = '_Atom_chem_shift.Atom_type'
190 SHIFT3 = '_Atom_chem_shift.Val'
191
192 def _create_shift(self, data):
193
194 try:
195 position = data[ChemShift3Reader.RANK3]
196 residue = data[ChemShift3Reader.RESIDUE3]
197 name = data[ChemShift3Reader.ATOM3]
198 element = data[ChemShift3Reader.ELEMENT3]
199 shift = data[ChemShift3Reader.SHIFT3]
200
201 except KeyError as ke:
202 raise ChemShiftFormatError("Required field {0} not found".format(str(ke)))
203 except ValueError as ve:
204 raise ChemShiftFormatError("Can't parse value: {0}".format(str(ve)))
205
206 return ChemShiftInfo(position, residue, name, element, shift)