20 Wrappers for the tmalign and tmscore utilities.
24 tmscore: Yang Zhang and Jeffrey Skolnick, Proteins 2004 57: 702-710
25 tmalign: Y. Zhang and J. Skolnick, Nucl. Acids Res. 2005 33, 2302-9
28 Authors: Pascal Benkert, Marco Biasini
31 import subprocess, os, tempfile, platform
33 from ost
import settings, io, geom, seq
35 def _SetupFiles(models, custom_chain_mapping = None):
37 tmp_dir_name=tempfile.mkdtemp()
39 for index, model
in enumerate(models):
40 for chain
in model.chains:
41 if len(chain.name) > 1:
44 for res
in chain.residues:
48 io.SavePDB(model, os.path.join(tmp_dir_name,
'model%02d.pdb' % (index+1)), dialect=dia)
49 if custom_chain_mapping
is not None:
50 with open(os.path.join(tmp_dir_name,
"custom_mapping.txt"),
'w')
as fh:
51 fh.write(
'\n'.join([f
"{mdl_ch}\t{ref_ch}" for ref_ch, mdl_ch
in custom_chain_mapping.items()]))
54 def _CleanupFiles(dir_name):
56 shutil.rmtree(dir_name)
58 def _ParseTmAlign(lines,lines_matrix):
59 info_line=lines[12].split(
',')
60 aln_length=int(info_line[0].split(
'=')[1].strip())
61 rmsd=float(info_line[1].split(
'=')[1].strip())
62 tm_score_swapped=float(lines[13].split(
'=')[1].split(
'(')[0].strip())
63 tm_score=float(lines[14].split(
'=')[1].split(
'(')[0].strip())
64 tf1=[float(i.strip())
for i
in lines_matrix[2].split()]
65 tf2=[float(i.strip())
for i
in lines_matrix[3].split()]
66 tf3=[float(i.strip())
for i
in lines_matrix[4].split()]
67 rot=
geom.Mat3(tf1[2], tf1[3], tf1[4], tf2[2], tf2[3],
68 tf2[4], tf3[2], tf3[3], tf3[4])
70 tf.PasteTranslation(
geom.Vec3(tf1[1], tf2[1], tf3[1]))
71 seq1 = seq.CreateSequence(
"1",lines[18].strip())
72 seq2 = seq.CreateSequence(
"2",lines[20].strip())
73 alignment = seq.CreateAlignment()
74 alignment.AddSequence(seq2)
75 alignment.AddSequence(seq1)
77 aln_length, tf, alignment)
79 def _ParseUSAlign(lines,lines_matrix):
84 tm_score_swapped =
None
89 mapping_data1 = list()
90 mapping_data2 = list()
95 if len(line.strip()) == 0:
98 aln_data.append(line.strip(
'*'))
99 elif line.startswith(
"Name of Structure_1:"):
100 tmp = [item.strip()
for item
in line.split()[3].split(
':')[1:]]
103 mapping_data1.append(item.split(
',')[1])
105 mapping_data1.append(
"")
106 elif line.startswith(
"Name of Structure_2:"):
107 tmp = [item.strip()
for item
in line.split()[3].split(
':')[1:]]
110 mapping_data2.append(item.split(
',')[1])
112 mapping_data2.append(
"")
113 elif line.startswith(
"Aligned length="):
114 data = [item.strip()
for item
in line.split(
',')]
116 if item.startswith(
"Aligned length="):
117 aligned_length = int(item.split(
"=")[1])
118 elif item.startswith(
"RMSD="):
119 rmsd = float(item.split(
"=")[1])
120 elif line.startswith(
"TM-score="):
121 if "(normalized by length of Structure_1" in line:
122 tm_score_swapped = float(line.split(
'(')[0].split(
'=')[1].strip())
123 elif "(normalized by length of Structure_2" in line:
124 tm_score = float(line.split(
'(')[0].split(
'=')[1].strip())
125 elif line.startswith(
"(\":\" denotes residue pairs of"):
128 assert(len(aln_data)==3)
129 aln_sequences1 = aln_data[0].split(
'*')
130 aln_sequences2 = aln_data[2].split(
'*')
134 ent1_mapped_chains = ost.StringList()
135 ent2_mapped_chains = ost.StringList()
136 assert(len(mapping_data1) == len(mapping_data2))
137 assert(len(aln_sequences1) == len(aln_sequences2))
138 assert(len(mapping_data1) == len(aln_sequences1))
139 for a, b, c, d
in zip(mapping_data1, mapping_data2,
140 aln_sequences1, aln_sequences2):
141 if len(a) > 0
and len(b) > 0:
142 ent1_mapped_chains.append(a)
143 ent2_mapped_chains.append(b)
144 assert(len(c) == len(d))
145 aln = seq.CreateAlignment()
146 aln.AddSequence(seq.CreateSequence(a, c))
147 aln.AddSequence(seq.CreateSequence(b, d))
151 tf1=[float(i.strip())
for i
in lines_matrix[2].split()[1:]]
152 tf2=[float(i.strip())
for i
in lines_matrix[3].split()[1:]]
153 tf3=[float(i.strip())
for i
in lines_matrix[4].split()[1:]]
154 mat =
geom.Mat4(tf1[1], tf1[2], tf1[3], tf1[0],
155 tf2[1], tf2[2], tf2[3], tf2[0],
156 tf3[1], tf3[2], tf3[3], tf3[0],
160 aligned_length, mat, alns,
161 ent1_mapped_chains, ent2_mapped_chains)
163 def _RunTmAlign(tmalign, tmp_dir):
164 model1_filename=os.path.join(tmp_dir,
'model01.pdb')
165 model2_filename=os.path.join(tmp_dir,
'model02.pdb')
166 if platform.system() ==
"Windows":
167 tmalign_path=settings.Locate(
'tmalign.exe', explicit_file_name=tmalign)
168 command=
"\"%s\" %s %s -m %s" %(os.path.normpath(tmalign_path), model1_filename, model2_filename, os.path.join(tmp_dir,
'matrix.txt'))
170 tmalign_path=settings.Locate(
'tmalign', explicit_file_name=tmalign)
171 command=
"\"%s\" \"%s\" \"%s\" -m \"%s\"" %(tmalign_path, model1_filename, model2_filename, os.path.join(tmp_dir,
'matrix.txt'))
172 ps=subprocess.Popen(command, shell=
True, stdout=subprocess.PIPE)
173 stdout,_=ps.communicate()
174 lines=stdout.decode().splitlines()
176 _CleanupFiles(tmp_dir)
177 raise RuntimeError(
"tmalign superposition failed")
178 matrix_file=open(os.path.join(tmp_dir,
'matrix.txt'))
179 lines_matrix=matrix_file.readlines()
181 return _ParseTmAlign(lines,lines_matrix)
183 def _RunUSAlign(usalign, tmp_dir):
184 model1_filename=os.path.join(tmp_dir,
'model01.pdb')
185 model2_filename=os.path.join(tmp_dir,
'model02.pdb')
186 mat_filename = os.path.join(tmp_dir,
"mat.txt")
187 usalign_path=settings.Locate(
'USalign', explicit_file_name=usalign)
188 command = f
"{usalign_path} {model1_filename} {model2_filename} -mm 1 -ter 0 -m {mat_filename}"
189 custom_mapping = os.path.join(tmp_dir,
"custom_mapping.txt")
190 if os.path.exists(custom_mapping):
191 command += f
" -chainmap {custom_mapping}"
192 ps=subprocess.Popen(command, shell=
True, stdout=subprocess.PIPE)
193 stdout,_=ps.communicate()
194 lines=stdout.decode().splitlines()
196 _CleanupFiles(tmp_dir)
197 raise RuntimeError(
"USalign superposition failed")
198 with open(mat_filename)
as fh:
199 lines_matrix = fh.readlines()
200 return _ParseUSAlign(lines,lines_matrix)
204 Holds the result of running TMscore
206 .. attribute:: rmsd_common
208 The RMSD of the common Calpha atoms of both structures
210 .. attribute:: rmsd_below_five
212 The RMSD of all Calpha atoms that can be superposed below five Angstroem
214 .. attribute:: tm_score
216 The TM-score of the structural superposition
218 .. attribute:: transform
220 The transform that superposes the model onto the reference structure.
222 :type: :class:`~ost.geom.Mat4`
224 .. attribute:: gdt_ha
226 The GDT_HA of the model to the reference structure.
228 .. attribute:: gdt_ts
230 The GDT_TS of the model to the reference structure.
234 gdt_ts, gdt_ha, rmsd_below_five, transform):
243 def _ParseTmScore(lines):
244 tf1=[float(i.strip())
for i
in lines[23].split()]
245 tf2=[float(i.strip())
for i
in lines[24].split()]
246 tf3=[float(i.strip())
for i
in lines[25].split()]
247 rot=
geom.Mat3(tf1[2], tf1[3], tf1[4], tf2[2], tf2[3],
248 tf2[4], tf3[2], tf3[3], tf3[4])
250 tf.PasteTranslation(
geom.Vec3(tf1[1], tf2[1], tf3[1]))
252 float(lines[16].split()[2].strip()),
253 float(lines[17].split()[1].strip()),
254 float(lines[18].split()[1].strip()),
255 float(lines[19].split()[1].strip()),
256 float(lines[27].split()[-1].strip()),
260 def _RunTmScore(tmscore, tmp_dir):
261 model1_filename=os.path.join(tmp_dir,
'model01.pdb')
262 model2_filename=os.path.join(tmp_dir,
'model02.pdb')
263 if platform.system() ==
"Windows":
264 tmscore_path=settings.Locate(
'tmscore.exe', explicit_file_name=tmscore)
265 command=
"\"%s\" %s %s" %(os.path.normpath(tmscore_path), model1_filename,
268 tmscore_path=settings.Locate(
'tmscore', explicit_file_name=tmscore)
269 command=
"\"%s\" \"%s\" \"%s\"" % (tmscore_path, model1_filename,
271 ps=subprocess.Popen(command, shell=
True, stdout=subprocess.PIPE)
272 stdout,_=ps.communicate()
273 lines=stdout.decode().splitlines()
275 _CleanupFiles(tmp_dir)
276 raise RuntimeError(
"tmscore superposition failed")
277 return _ParseTmScore(lines)
282 Performs a sequence independent superposition of model1 onto model2, the
286 :param model1: The model structure. If the superposition is successful, will
287 be superposed onto the reference structure
288 :type model1: :class:`~ost.mol.EntityView` or :class:`~ost.mol.EntityHandle`
289 :param model2: The reference structure
290 :type model2: :class:`~ost.mol.EntityView` or :class:`~ost.mol.EntityHandle`
291 :param tmalign: If not None, the path to the tmalign executable.
292 :returns: The result of the tmscore superposition
293 :rtype: :class:`ost.bindings.TMAlignResult`
295 :raises: :class:`~ost.settings.FileNotFound` if tmalign could not be located.
296 :raises: :class:`RuntimeError` if the superposition failed
298 tmp_dir_name=_SetupFiles((model1, model2))
299 result=_RunTmAlign(tmalign, tmp_dir_name)
300 model1.handle.EditXCS().ApplyTransform(result.transform)
301 _CleanupFiles(tmp_dir_name)
306 Performs a sequence dependent superposition of model1 onto model2,
309 :param model1: The model structure. If the superposition is successful, will
310 be superposed onto the reference structure
311 :type model1: :class:`~ost.mol.EntityView` or :class:`~ost.mol.EntityHandle`
312 :param model2: The reference structure
313 :type model2: :class:`~ost.mol.EntityView` or :class:`~ost.mol.EntityHandle`
314 :param tmscore: If not None, the path to the tmscore executable.
315 :returns: The result of the tmscore superposition
316 :rtype: :class:`TMScoreResult`
318 :raises: :class:`~ost.settings.FileNotFound` if tmalign could not be located.
319 :raises: :class:`RuntimeError` if the superposition failed
321 tmp_dir_name=_SetupFiles((model1, model2))
322 result=_RunTmScore(tmscore, tmp_dir_name)
323 model1.handle.EditXCS().ApplyTransform(result.transform)
324 _CleanupFiles(tmp_dir_name)
328 def USAlign(model1, model2, usalign=None, custom_chain_mapping=None):
330 Performs a sequence independent superposition of model1 onto model2, the
331 reference. Can deal with multimeric complexes and RNA.
333 Creates temporary model files on disk and runs USalign with:
334 ``USalign model1.pdb model2.pdb -mm 1 -ter 0 -m rotmat.txt``
336 :param model1: The model structure. If the superposition is successful, will
337 be superposed onto the reference structure
338 :type model1: :class:`~ost.mol.EntityView` or :class:`~ost.mol.EntityHandle`
339 :param model2: The reference structure
340 :type model2: :class:`~ost.mol.EntityView` or :class:`~ost.mol.EntityHandle`
341 :param usalign: If not None, the path to the USalign executable. Searches
342 for executable with name ``USalign`` in PATH if not given.
343 :param custom_chain_mapping: Custom chain mapping that is passed as -chainmap
344 to USalign executable. Raises an error is this
345 is not supported by the USalign executable you're
346 using (introduced in July 2023).
347 It's a dict with reference chain names as key
348 (model2) and model chain names as values
350 :type custom_chain_mapping: :class:`dict`
351 :returns: The result of the superposition
352 :rtype: :class:`ost.bindings.MMAlignResult`
354 :raises: :class:`~ost.settings.FileNotFound` if executable could not be located.
355 :raises: :class:`RuntimeError` if the superposition failed
357 tmp_dir_name=_SetupFiles((model1, model2),
358 custom_chain_mapping=custom_chain_mapping)
359 result=_RunUSAlign(usalign, tmp_dir_name)
360 model1.handle.EditXCS().ApplyTransform(result.transform)
361 _CleanupFiles(tmp_dir_name)
Three dimensional vector class, using Real precision.
std::vector< AlignmentHandle > AlignmentList