DCDConverter.py 12.7 KB
Newer Older
1
2
3
4
5
#MDANSE : Molecular Dynamics Analysis for Neutron Scattering Experiments
#------------------------------------------------------------------------------------------
#Copyright (C)
#2015- Eric C. Pellegrini Institut Laue-Langevin
#BP 156
6
7
#71 avenue des Martyrs
#38000 Grenoble Cedex 9
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
#France
#pellegrini[at]ill.fr
#goret[at]ill.fr
#aoun[at]ill.fr
#
#This library is free software; you can redistribute it and/or
#modify it under the terms of the GNU Lesser General Public
#License as published by the Free Software Foundation; either
#version 2.1 of the License, or (at your option) any later version.
#
#This library is distributed in the hope that it will be useful,
#but WITHOUT ANY WARRANTY; without even the implied warranty of
#MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
#Lesser General Public License for more details.
#
#You should have received a copy of the GNU Lesser General Public
#License along with this library; if not, write to the Free Software
#Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA

''' 
Created on Apr 10, 2015

30
:author: Eric C. Pellegrini
31
32
'''

33
import collections
34
import os
35
36
37
38
39
40
41
42
43
44
45
import struct

import numpy

from MMTK import Units
from MMTK.ParticleProperties import Configuration
from MMTK.PDB import PDBConfiguration
from MMTK.Trajectory import Trajectory, SnapshotGenerator, TrajectoryOutput
from MMTK.Universe import InfiniteUniverse, ParallelepipedicPeriodicUniverse

from MDANSE.Core.Error import Error
46
from MDANSE.Framework.Jobs.Converter import Converter
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
from MDANSE.Mathematics.Geometry import get_basis_vectors_from_cell_parameters
from MDANSE.MolecularDynamics.Trajectory import resolve_undefined_molecules_name

PI_2 = 0.5*numpy.pi
RECSCALE32BIT = 1
RECSCALE64BIT = 2

class DCDFileError(Error):
    pass
  
class ByteOrderError(Error):
    pass

class InputOutputError(Error):
    pass

class EndOfFile(Error):
    pass

class FortranBinaryFileError(Error):
    pass

def get_byte_order(filename):

    # Identity the byte order of the file by trial-and-error
    byteOrder = None

    # The DCD file is opened for reading in binary mode.
    data = file(filename, 'rb').read(4)

    # Check for low and big endianness byte orders.
    for order in ['<', '>']:
        reclen = struct.unpack(order + 'i', data)[0]
        if reclen == 84:
            byteOrder = order
            break

        if byteOrder is None:
            raise ByteOrderError("Invalid byte order. %s is not a valid DCD file" % filename)
                
    return byteOrder

class FortranBinaryFile(object):
    """Sets up a Fortran binary file reader. 

    @note: written by Konrad Hinsen.
    """
    def __init__(self, filename):
        """The constructor.

        @param filename: the input file.
        @type filename: string.

        @param byte_order: the byte order to read the binary file.
        @type byte_order: string being one '@', '=', '<', '>' or '!'.
        """
        self.file = file(filename, 'rb')
        self.byteOrder = get_byte_order(filename)

    def __iter__(self):
        return self

    def next_record(self):
        data = self.file.read(struct.calcsize("i"))
        if not data:
            raise StopIteration
        reclen = struct.unpack(self.byteOrder + 'i', data)[0]
        data = self.file.read(reclen)
        reclen2 = struct.unpack(self.byteOrder + 'i', self.file.read(struct.calcsize("i")))[0]
        if reclen != reclen2:
            FortranBinaryFileError("Invalid block")
            
        return data

    def skip_record(self):
        data = self.file.read(struct.calcsize("i"))
        reclen = struct.unpack(self.byteOrder + 'i', data)[0]
        self.file.seek(reclen, 1)
        reclen2 = struct.unpack(self.byteOrder + 'i', self.file.read(4))[0]
        assert reclen==reclen2

    def get_record(self, fmt, repeat = False):
        """Reads a record of the binary file.

        @param format: the format corresponding to the binray structure to read.
        @type format: string.        

        @param repeat: if True, will repeat the reading.
        @type repeat: bool.        
        """

        try:
            data = self.next_record()
        except StopIteration:
            raise EndOfFile()
        if repeat:
            unit = struct.calcsize(self.byteOrder + fmt)
            assert len(data) % unit == 0
            fmt = (len(data)/unit) * fmt
        try:
            return struct.unpack(self.byteOrder + fmt, data)
        except:
            raise                
  
class DCDFile(FortranBinaryFile, dict):
        
    def __init__(self, filename):
        
        FortranBinaryFile.__init__(self, filename)
                        
        self['filename'] = filename
                                        
        self.read_header()
        
    def read_header(self):
        
        # Read a block
        data = self.next_record()
                
        if data[:4] != 'CORD':
            raise DCDFileError("Unrecognized DCD format")

        temp = struct.unpack(self.byteOrder + '20i', data[4:])
        
        self['charmm'] = temp[-1]
                
        if self['charmm']:
            temp = struct.unpack(self.byteOrder + '9if10i', data[4:])
        else:
            temp = struct.unpack(self.byteOrder + '9id9i', data[4:])

        # Store the number of sets of coordinates
        self['nset'] = self['n_frames'] = temp[0]
        
        # Store the starting time step
        self['istart'] = temp[1]
        
        # Store the number of timesteps between dcd saves
        self['nsavc'] = temp[2]
        
        # Stores the number of fixed atoms
        self['namnf'] = temp[8]

        # Stop if there are fixed atoms.
        if self['namnf'] > 0:
            raise DCDFileError('Can not handle fixed atoms yet.')
                                        
        self['delta'] = temp[9]
                                            
        self["time_step"] = self['nsavc']*self['delta']*Units.akma_time           

        self['has_pbc_data'] = temp[10]

        self['has_4d'] = temp[11]
        
        # Read a block
        data = self.next_record()
                
        nLines = struct.unpack(self.byteOrder + b'I', data[0:4])[0]
            
        self["title"] = []
        for i in range(nLines):                
            temp = struct.unpack(self.byteOrder + '80c', data[4+80*i:4+80*(i+1)])
            self["title"].append("".join(temp).strip())
        
        self["title"] = "\n".join(self["title"])
        
        # Read a block
        data = self.next_record()
        
        # Read the number of atoms.
        self['natoms'] = struct.unpack(self.byteOrder + b'I', data)[0]
                                                            
    def read_step(self):
        """
        Reads a frame of the DCD file.
        """
                
        if self['has_pbc_data']:
            unitCell = numpy.array(self.get_record('6d'), dtype = numpy.float64)
            unitCell = unitCell[[0,2,5,1,3,4]]
            unitCell[0:3] *= Units.Ang
            # This file was generated by CHARMM, or by NAMD > 2.5, with the angle
            # cosines of the periodic cell angles written to the DCD file.       
            # This formulation improves rounding behavior for orthogonal cells   
232
233
            # so that the angles end up at precisely 90 degrees, unlike acos(). 
            # See https://github.com/MDAnalysis/mdanalysis/wiki/FileFormats for info           
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
            if numpy.all(abs(unitCell[3:]) <= 1):
                unitCell[3:] = PI_2 - numpy.arcsin(unitCell[3:])
            else:
                # assume the angles are stored in degrees (NAMD <= 2.5)
                unitCell[3:] *= Units.deg
            

        else:
            unitCell = None

        fmt = '%df' % self['natoms']
        config = numpy.empty((self["natoms"],3),dtype=numpy.float64)
        config[:,0] = numpy.array(self.get_record(fmt), dtype=numpy.float64)
        config[:,1] = numpy.array(self.get_record(fmt), dtype=numpy.float64)
        config[:,2] = numpy.array(self.get_record(fmt), dtype=numpy.float64)
        config *= Units.Ang
        
        if self['has_4d']:
            self.skip_record()
            
        return unitCell, config

    def skip_step(self):
        """Skips a frame of the DCD file.
        """
        nrecords = 3
        if self['has_pbc_data']:
            nrecords += 1
        if self['has_4d']:
            nrecords += 1
        for _ in range(nrecords):
            self['binary'].skip_record()

    def __iter__(self):
        return self

    def next_step(self):
        try:
            return self.read_step()
        except EndOfFile:
            raise StopIteration

276
class DCDConverter(Converter):
277
278
279
    """
    Converts a DCD trajectory to a MMTK trajectory.
    """
280
281
    
    settings = collections.OrderedDict()
282
283
    settings['pdb_file'] = ('input_file',{'default':os.path.join('..','..','..','Data','Trajectories','CHARMM','2vb1.pdb')})
    settings['dcd_file'] = ('input_file',{'default':os.path.join('..','..','..','Data','Trajectories','CHARMM','2vb1.dcd')})
284
    settings['time_step'] = ('float', {'default':1.0,'label':"Time step (ps)"})    
285
    settings['fold'] = ('boolean', {'default':False,'label':"Fold coordinates in to box"})    
286
    settings['output_files'] = ('output_files', {'formats':["netcdf"]})
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317

    def initialize(self):
        """
        Initialize the input parameters and analysis self variables
        """

        self.configuration["dcd_file"]["instance"] = DCDFile(self.configuration["dcd_file"]['filename'])

        # The number of steps of the analysis.
        self.numberOfSteps = self.configuration['dcd_file']['instance']['n_frames']
 
        # Create all objects from the PDB file.  
        conf = PDBConfiguration(self.configuration['pdb_file']['filename'])

        # Creates a collection of all the chemical objects stored in the PDB file
        molecules = conf.createAll()
                        
        # If the input trajectory has PBC create a periodic universe.
        if self.configuration['dcd_file']['instance']['has_pbc_data']:
            self._universe = ParallelepipedicPeriodicUniverse()
            
        # Otherwise create an infinite universe.
        else:
            self._universe = InfiniteUniverse()
                    
        # The chemical objects found in the PDB file introduced into the universe.
        self._universe.addObject(molecules)

        resolve_undefined_molecules_name(self._universe)
        
        # A MMTK trajectory is opened for writing.
318
        self._trajectory = Trajectory(self._universe, self.configuration['output_files']['files'][0], mode='w')
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
        
        # A frame generator is created.        
        self._snapshot = SnapshotGenerator(self._universe, actions=[TrajectoryOutput(self._trajectory, ["all"], 0, None, 1)])

    def run_step(self, index):
        """
        Runs a single step of the job.\n
 
        :Parameters:
            #. index (int): The index of the step.
        :Returns:
            #. index (int): The index of the step. 
        """
                        
        # The x, y and z values of the current frame.
        unitCell, config = self.configuration["dcd_file"]["instance"].read_step()
        
        conf = Configuration(self._universe,config)
        
        # If the universe is periodic set its shape with the current dimensions of the unit cell.
        if self._universe.is_periodic:
            self._universe.setShape(get_basis_vectors_from_cell_parameters(unitCell))
        
        self._universe.setConfiguration(conf)
        
        if self.configuration['fold']["value"]:        
            self._universe.foldCoordinatesIntoBox()
                                                   
        # The current time.
348
        t = (index+1)*self.configuration["time_step"]["value"]
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370

        # Store a snapshot of the current configuration in the output trajectory.
        self._snapshot(data={'time': t})
                                        
        return index, None

    def combine(self, index, x):
        """
        Combines returned results of run_step.\n
        :Parameters:
            #. index (int): The index of the step.\n
            #. x (any): The returned result(s) of run_step
        """   
        
        pass
    
    def finalize(self):
        """
        Finalizes the calculations (e.g. averaging the total term, output files creations ...).
        """ 

        # Close the output trajectory.
371
        self._trajectory.close()