1 """@namespace IMP.pmi.output
2 Classes for writing output files and processing them.
5 from __future__
import print_function, division
24 import cPickle
as pickle
29 class _ChainIDs(object):
30 """Map indices to multi-character chain IDs.
31 We label the first 26 chains A-Z, then we move to two-letter
32 chain IDs: AA through AZ, then BA through BZ, through to ZZ.
33 This continues with longer chain IDs."""
34 def __getitem__(self, ind):
35 chars = string.ascii_uppercase
39 ids.append(chars[ind % lc])
41 ids.append(chars[ind])
42 return "".join(reversed(ids))
46 """Base class for capturing a modeling protocol.
47 Unlike simple output of model coordinates, a complete
48 protocol includes the input data used, details on the restraints,
49 sampling, and clustering, as well as output models.
50 Use via IMP.pmi.topology.System.add_protocol_output().
52 @see IMP.pmi.mmcif.ProtocolOutput for a concrete subclass that outputs
60 if isinstance(elt, (tuple, list)):
61 for elt2
in _flatten(elt):
67 def _disambiguate_chain(chid, seen_chains):
68 """Make sure that the chain ID is unique; warn and correct if it isn't"""
73 if chid
in seen_chains:
74 warnings.warn(
"Duplicate chain ID '%s' encountered" % chid,
77 for suffix
in itertools.count(1):
78 new_chid = chid +
"%d" % suffix
79 if new_chid
not in seen_chains:
80 seen_chains.add(new_chid)
86 def _write_pdb_internal(flpdb, particle_infos_for_pdb, geometric_center,
87 write_all_residues_per_bead):
88 for n, tupl
in enumerate(particle_infos_for_pdb):
89 (xyz, atom_type, residue_type,
90 chain_id, residue_index, all_indexes, radius) = tupl
92 atom_type = IMP.atom.AT_CA
93 if write_all_residues_per_bead
and all_indexes
is not None:
94 for residue_number
in all_indexes:
96 IMP.atom.get_pdb_string((xyz[0] - geometric_center[0],
97 xyz[1] - geometric_center[1],
98 xyz[2] - geometric_center[2]),
99 n+1, atom_type, residue_type,
100 chain_id[:1], residue_number,
' ',
104 IMP.atom.get_pdb_string((xyz[0] - geometric_center[0],
105 xyz[1] - geometric_center[1],
106 xyz[2] - geometric_center[2]),
107 n+1, atom_type, residue_type,
108 chain_id[:1], residue_index,
' ',
110 flpdb.write(
"ENDMDL\n")
113 _Entity = collections.namedtuple(
'_Entity', (
'id',
'seq'))
114 _ChainInfo = collections.namedtuple(
'_ChainInfo', (
'entity',
'name'))
117 def _get_chain_info(chains, root_hier):
121 for mol
in IMP.atom.get_by_type(root_hier, IMP.atom.MOLECULE_TYPE):
123 chain_id = chains[molname]
125 seq = chain.get_sequence()
126 if seq
not in entities:
127 entities[seq] = e = _Entity(id=len(entities)+1, seq=seq)
128 all_entities.append(e)
129 entity = entities[seq]
130 info = _ChainInfo(entity=entity, name=molname)
131 chain_info[chain_id] = info
132 return chain_info, all_entities
135 def _write_mmcif_internal(flpdb, particle_infos_for_pdb, geometric_center,
136 write_all_residues_per_bead, chains, root_hier):
138 chain_info, entities = _get_chain_info(chains, root_hier)
140 writer = ihm.format.CifWriter(flpdb)
141 writer.start_block(
'model')
142 with writer.category(
"_entry")
as lp:
145 with writer.loop(
"_entity", [
"id",
"type"])
as lp:
147 lp.write(id=e.id, type=
"polymer")
149 with writer.loop(
"_entity_poly",
150 [
"entity_id",
"pdbx_seq_one_letter_code"])
as lp:
152 lp.write(entity_id=e.id, pdbx_seq_one_letter_code=e.seq)
154 with writer.loop(
"_struct_asym", [
"id",
"entity_id",
"details"])
as lp:
156 for chid
in sorted(chains.values(), key=
lambda x: (len(x.strip()), x)):
157 ci = chain_info[chid]
158 lp.write(id=chid, entity_id=ci.entity.id, details=ci.name)
160 with writer.loop(
"_atom_site",
161 [
"group_PDB",
"type_symbol",
"label_atom_id",
162 "label_comp_id",
"label_asym_id",
"label_seq_id",
164 "Cartn_x",
"Cartn_y",
"Cartn_z",
"label_entity_id",
165 "pdbx_pdb_model_num",
168 for n, tupl
in enumerate(particle_infos_for_pdb):
169 (xyz, atom_type, residue_type,
170 chain_id, residue_index, all_indexes, radius) = tupl
171 ci = chain_info[chain_id]
172 if atom_type
is None:
173 atom_type = IMP.atom.AT_CA
174 c = (xyz[0] - geometric_center[0],
175 xyz[1] - geometric_center[1],
176 xyz[2] - geometric_center[2])
177 if write_all_residues_per_bead
and all_indexes
is not None:
178 for residue_number
in all_indexes:
179 lp.write(group_PDB=
'ATOM',
181 label_atom_id=atom_type.get_string(),
182 label_comp_id=residue_type.get_string(),
183 label_asym_id=chain_id,
184 label_seq_id=residue_index,
185 auth_seq_id=residue_index, Cartn_x=c[0],
186 Cartn_y=c[1], Cartn_z=c[2], id=ordinal,
187 pdbx_pdb_model_num=1,
188 label_entity_id=ci.entity.id)
191 lp.write(group_PDB=
'ATOM', type_symbol=
'C',
192 label_atom_id=atom_type.get_string(),
193 label_comp_id=residue_type.get_string(),
194 label_asym_id=chain_id,
195 label_seq_id=residue_index,
196 auth_seq_id=residue_index, Cartn_x=c[0],
197 Cartn_y=c[1], Cartn_z=c[2], id=ordinal,
198 pdbx_pdb_model_num=1,
199 label_entity_id=ci.entity.id)
204 """Class for easy writing of PDBs, RMFs, and stat files
206 @note Model should be updated prior to writing outputs.
208 def __init__(self, ascii=True, atomistic=False):
209 self.dictionary_pdbs = {}
211 self.dictionary_rmfs = {}
212 self.dictionary_stats = {}
213 self.dictionary_stats2 = {}
214 self.best_score_list =
None
215 self.nbestscoring =
None
217 self.replica_exchange =
False
222 self.chainids =
"ABCDEFGHIJKLMNOPQRSTUVWXYZ" \
223 "abcdefghijklmnopqrstuvwxyz0123456789"
225 self.multi_chainids = _ChainIDs()
227 self.particle_infos_for_pdb = {}
228 self.atomistic = atomistic
231 """Get a list of all PDB files being output by this instance"""
232 return list(self.dictionary_pdbs.keys())
234 def get_rmf_names(self):
235 return list(self.dictionary_rmfs.keys())
237 def get_stat_names(self):
238 return list(self.dictionary_stats.keys())
240 def _init_dictchain(self, name, prot, multichar_chain=False, mmcif=False):
241 self.dictchain[name] = {}
245 self.atomistic =
True
246 for n, mol
in enumerate(IMP.atom.get_by_type(
247 prot, IMP.atom.MOLECULE_TYPE)):
249 if not mmcif
and len(chid) > 1:
251 "The system contains at least one chain ID (%s) that "
252 "is more than 1 character long; this cannot be "
253 "represented in PDB. Either write mmCIF files "
254 "instead, or assign 1-character IDs to all chains "
255 "(this can be done with the `chain_ids` argument to "
256 "BuildSystem.add_state())." % chid)
257 chid = _disambiguate_chain(chid, seen_chains)
259 self.dictchain[name][molname] = chid
263 @param name The PDB filename
264 @param prot The hierarchy to write to this pdb file
265 @param mmcif If True, write PDBs in mmCIF format
266 @note if the PDB name is 'System' then will use Selection
269 flpdb = open(name,
'w')
271 self.dictionary_pdbs[name] = prot
272 self._pdb_mmcif[name] = mmcif
273 self._init_dictchain(name, prot, mmcif=mmcif)
275 def write_psf(self, filename, name):
276 flpsf = open(filename,
'w')
277 flpsf.write(
"PSF CMAP CHEQ" +
"\n")
278 index_residue_pair_list = {}
279 (particle_infos_for_pdb, geometric_center) = \
280 self.get_particle_infos_for_pdb_writing(name)
281 nparticles = len(particle_infos_for_pdb)
282 flpsf.write(str(nparticles) +
" !NATOM" +
"\n")
283 for n, p
in enumerate(particle_infos_for_pdb):
288 flpsf.write(
'{0:8d}{1:1s}{2:4s}{3:1s}{4:4s}{5:1s}{6:4s}{7:1s}'
289 '{8:4s}{9:1s}{10:4s}{11:14.6f}{12:14.6f}{13:8d}'
290 '{14:14.6f}{15:14.6f}'.format(
291 atom_index,
" ", chain,
" ", str(resid),
" ",
292 '"'+residue_type.get_string()+
'"',
" ",
"C",
293 " ",
"C", 1.0, 0.0, 0, 0.0, 0.0))
295 if chain
not in index_residue_pair_list:
296 index_residue_pair_list[chain] = [(atom_index, resid)]
298 index_residue_pair_list[chain].append((atom_index, resid))
302 for chain
in sorted(index_residue_pair_list.keys()):
304 ls = index_residue_pair_list[chain]
306 ls = sorted(ls, key=
lambda tup: tup[1])
308 indexes = [x[0]
for x
in ls]
311 indexes, lmin=2, lmax=2))
312 nbonds = len(indexes_pairs)
313 flpsf.write(str(nbonds)+
" !NBOND: bonds"+
"\n")
316 for i
in range(0, len(indexes_pairs), 4):
317 for bond
in indexes_pairs[i:i+4]:
318 flpsf.write(
'{0:8d}{1:8d}'.format(*bond))
321 del particle_infos_for_pdb
324 def write_pdb(self, name, appendmode=True,
325 translate_to_geometric_center=
False,
326 write_all_residues_per_bead=
False):
328 (particle_infos_for_pdb,
329 geometric_center) = self.get_particle_infos_for_pdb_writing(name)
331 if not translate_to_geometric_center:
332 geometric_center = (0, 0, 0)
334 filemode =
'a' if appendmode
else 'w'
335 with open(name, filemode)
as flpdb:
336 if self._pdb_mmcif[name]:
337 _write_mmcif_internal(flpdb, particle_infos_for_pdb,
339 write_all_residues_per_bead,
340 self.dictchain[name],
341 self.dictionary_pdbs[name])
343 _write_pdb_internal(flpdb, particle_infos_for_pdb,
345 write_all_residues_per_bead)
348 """Get the protein name from the particle.
349 This is done by traversing the hierarchy."""
352 def get_particle_infos_for_pdb_writing(self, name):
362 particle_infos_for_pdb = []
364 geometric_center = [0, 0, 0]
369 and self.dictionary_pdbs[name].get_number_of_children() == 0):
373 ps = sel.get_selected_particles()
375 for n, p
in enumerate(ps):
378 if protname
not in resindexes_dict:
379 resindexes_dict[protname] = []
383 rt = residue.get_residue_type()
384 resind = residue.get_index()
388 geometric_center[0] += xyz[0]
389 geometric_center[1] += xyz[1]
390 geometric_center[2] += xyz[2]
392 particle_infos_for_pdb.append(
393 (xyz, atomtype, rt, self.dictchain[name][protname],
394 resind,
None, radius))
395 resindexes_dict[protname].append(resind)
400 resind = residue.get_index()
403 if resind
in resindexes_dict[protname]:
406 resindexes_dict[protname].append(resind)
407 rt = residue.get_residue_type()
410 geometric_center[0] += xyz[0]
411 geometric_center[1] += xyz[1]
412 geometric_center[2] += xyz[2]
414 particle_infos_for_pdb.append(
415 (xyz,
None, rt, self.dictchain[name][protname], resind,
420 resind = resindexes[len(resindexes) // 2]
421 if resind
in resindexes_dict[protname]:
424 resindexes_dict[protname].append(resind)
428 geometric_center[0] += xyz[0]
429 geometric_center[1] += xyz[1]
430 geometric_center[2] += xyz[2]
432 particle_infos_for_pdb.append(
433 (xyz,
None, rt, self.dictchain[name][protname], resind,
440 if len(resindexes) > 0:
441 resind = resindexes[len(resindexes) // 2]
444 geometric_center[0] += xyz[0]
445 geometric_center[1] += xyz[1]
446 geometric_center[2] += xyz[2]
448 particle_infos_for_pdb.append(
449 (xyz,
None, rt, self.dictchain[name][protname],
450 resind, resindexes, radius))
453 geometric_center = (geometric_center[0] / atom_count,
454 geometric_center[1] / atom_count,
455 geometric_center[2] / atom_count)
459 particle_infos_for_pdb = sorted(particle_infos_for_pdb,
460 key=
lambda x: (len(x[3]), x[3], x[4]))
462 return (particle_infos_for_pdb, geometric_center)
464 def write_pdbs(self, appendmode=True, mmcif=False):
465 for pdb
in self.dictionary_pdbs.keys():
466 self.write_pdb(pdb, appendmode)
469 replica_exchange=
False, mmcif=
False,
470 best_score_file=
'best.scores.rex.py'):
471 """Prepare for writing best-scoring PDBs (or mmCIFs) for a
474 @param prefix Initial part of each PDB filename (e.g. 'model').
475 @param prot The top-level Hierarchy to output.
476 @param nbestscoring The number of best-scoring files to output.
477 @param replica_exchange Whether to combine best scores from a
478 replica exchange run.
479 @param mmcif If True, output models in mmCIF format. If False
480 (the default) output in legacy PDB format.
481 @param best_score_file The filename to use for replica
485 self._pdb_best_scoring_mmcif = mmcif
486 fileext =
'.cif' if mmcif
else '.pdb'
487 self.prefixes.append(prefix)
488 self.replica_exchange = replica_exchange
489 if not self.replica_exchange:
493 self.best_score_list = []
497 self.best_score_file_name = best_score_file
498 self.best_score_list = []
499 with open(self.best_score_file_name,
"w")
as best_score_file:
500 best_score_file.write(
501 "self.best_score_list=" + str(self.best_score_list) +
"\n")
503 self.nbestscoring = nbestscoring
504 for i
in range(self.nbestscoring):
505 name = prefix +
"." + str(i) + fileext
506 flpdb = open(name,
'w')
508 self.dictionary_pdbs[name] = prot
509 self._pdb_mmcif[name] = mmcif
510 self._init_dictchain(name, prot, mmcif=mmcif)
512 def write_pdb_best_scoring(self, score):
513 if self.nbestscoring
is None:
514 print(
"Output.write_pdb_best_scoring: init_pdb_best_scoring "
517 mmcif = self._pdb_best_scoring_mmcif
518 fileext =
'.cif' if mmcif
else '.pdb'
520 if self.replica_exchange:
522 with open(self.best_score_file_name)
as fh:
523 self.best_score_list = ast.literal_eval(
524 fh.read().split(
'=')[1])
526 if len(self.best_score_list) < self.nbestscoring:
527 self.best_score_list.append(score)
528 self.best_score_list.sort()
529 index = self.best_score_list.index(score)
530 for prefix
in self.prefixes:
531 for i
in range(len(self.best_score_list) - 2, index - 1, -1):
532 oldname = prefix +
"." + str(i) + fileext
533 newname = prefix +
"." + str(i + 1) + fileext
535 if os.path.exists(newname):
537 os.rename(oldname, newname)
538 filetoadd = prefix +
"." + str(index) + fileext
539 self.write_pdb(filetoadd, appendmode=
False)
542 if score < self.best_score_list[-1]:
543 self.best_score_list.append(score)
544 self.best_score_list.sort()
545 self.best_score_list.pop(-1)
546 index = self.best_score_list.index(score)
547 for prefix
in self.prefixes:
548 for i
in range(len(self.best_score_list) - 1,
550 oldname = prefix +
"." + str(i) + fileext
551 newname = prefix +
"." + str(i + 1) + fileext
552 os.rename(oldname, newname)
553 filenametoremove = prefix + \
554 "." + str(self.nbestscoring) + fileext
555 os.remove(filenametoremove)
556 filetoadd = prefix +
"." + str(index) + fileext
557 self.write_pdb(filetoadd, appendmode=
False)
559 if self.replica_exchange:
561 with open(self.best_score_file_name,
"w")
as best_score_file:
562 best_score_file.write(
563 "self.best_score_list=" + str(self.best_score_list) +
'\n')
565 def init_rmf(self, name, hierarchies, rs=None, geometries=None,
568 Initialize an RMF file
570 @param name the name of the RMF file
571 @param hierarchies the hierarchies to be included (it is a list)
572 @param rs optional, the restraint sets (it is a list)
573 @param geometries optional, the geometries (it is a list)
574 @param listofobjects optional, the list of objects for the stat
577 rh = RMF.create_rmf_file(name)
580 outputkey_rmfkey =
None
584 if geometries
is not None:
586 if listofobjects
is not None:
587 cat = rh.get_category(
"stat")
588 outputkey_rmfkey = {}
589 for o
in listofobjects:
590 if "get_output" not in dir(o):
592 "Output: object %s doesn't have get_output() method"
594 output = o.get_output()
595 for outputkey
in output:
596 rmftag = RMF.string_tag
597 if isinstance(output[outputkey], float):
598 rmftag = RMF.float_tag
599 elif isinstance(output[outputkey], int):
601 elif isinstance(output[outputkey], str):
602 rmftag = RMF.string_tag
604 rmftag = RMF.string_tag
605 rmfkey = rh.get_key(cat, outputkey, rmftag)
606 outputkey_rmfkey[outputkey] = rmfkey
607 outputkey_rmfkey[
"rmf_file"] = \
608 rh.get_key(cat,
"rmf_file", RMF.string_tag)
609 outputkey_rmfkey[
"rmf_frame_index"] = \
610 rh.get_key(cat,
"rmf_frame_index", RMF.int_tag)
612 self.dictionary_rmfs[name] = (rh, cat, outputkey_rmfkey, listofobjects)
614 def add_restraints_to_rmf(self, name, objectlist):
615 for o
in _flatten(objectlist):
617 rs = o.get_restraint_for_rmf()
618 if not isinstance(rs, (list, tuple)):
621 rs = [o.get_restraint()]
623 self.dictionary_rmfs[name][0], rs)
625 def add_geometries_to_rmf(self, name, objectlist):
627 geos = o.get_geometries()
630 def add_particle_pair_from_restraints_to_rmf(self, name, objectlist):
633 pps = o.get_particle_pairs()
636 self.dictionary_rmfs[name][0],
639 def write_rmf(self, name):
641 if self.dictionary_rmfs[name][1]
is not None:
642 outputkey_rmfkey = self.dictionary_rmfs[name][2]
643 listofobjects = self.dictionary_rmfs[name][3]
644 for o
in listofobjects:
645 output = o.get_output()
646 for outputkey
in output:
647 rmfkey = outputkey_rmfkey[outputkey]
649 n = self.dictionary_rmfs[name][0].get_root_node()
650 n.set_value(rmfkey, output[outputkey])
651 except NotImplementedError:
653 rmfkey = outputkey_rmfkey[
"rmf_file"]
654 self.dictionary_rmfs[name][0].get_root_node().set_value(
656 rmfkey = outputkey_rmfkey[
"rmf_frame_index"]
658 self.dictionary_rmfs[name][0].get_root_node().set_value(
660 self.dictionary_rmfs[name][0].flush()
662 def close_rmf(self, name):
663 rh = self.dictionary_rmfs[name][0]
664 del self.dictionary_rmfs[name]
667 def write_rmfs(self):
668 for rmfinfo
in self.dictionary_rmfs.keys():
669 self.write_rmf(rmfinfo[0])
671 def init_stat(self, name, listofobjects):
673 flstat = open(name,
'w')
676 flstat = open(name,
'wb')
680 for o
in listofobjects:
681 if "get_output" not in dir(o):
683 "Output: object %s doesn't have get_output() method"
685 self.dictionary_stats[name] = listofobjects
687 def set_output_entry(self, key, value):
688 self.initoutput.update({key: value})
690 def write_stat(self, name, appendmode=True):
691 output = self.initoutput
692 for obj
in self.dictionary_stats[name]:
695 dfiltered = dict((k, v)
for k, v
in d.items()
if k[0] !=
"_")
696 output.update(dfiltered)
704 flstat = open(name, writeflag)
705 flstat.write(
"%s \n" % output)
708 flstat = open(name, writeflag +
'b')
709 pickle.dump(output, flstat, 2)
712 def write_stats(self):
713 for stat
in self.dictionary_stats.keys():
714 self.write_stat(stat)
716 def get_stat(self, name):
718 for obj
in self.dictionary_stats[name]:
719 output.update(obj.get_output())
722 def write_test(self, name, listofobjects):
723 flstat = open(name,
'w')
724 output = self.initoutput
725 for o
in listofobjects:
726 if "get_test_output" not in dir(o)
and "get_output" not in dir(o):
728 "Output: object %s doesn't have get_output() or "
729 "get_test_output() method" % str(o))
730 self.dictionary_stats[name] = listofobjects
732 for obj
in self.dictionary_stats[name]:
734 d = obj.get_test_output()
738 dfiltered = dict((k, v)
for k, v
in d.items()
if k[0] !=
"_")
739 output.update(dfiltered)
740 flstat.write(
"%s \n" % output)
743 def test(self, name, listofobjects, tolerance=1e-5):
744 output = self.initoutput
745 for o
in listofobjects:
746 if "get_test_output" not in dir(o)
and "get_output" not in dir(o):
748 "Output: object %s doesn't have get_output() or "
749 "get_test_output() method" % str(o))
750 for obj
in listofobjects:
752 output.update(obj.get_test_output())
754 output.update(obj.get_output())
756 flstat = open(name,
'r')
760 test_dict = ast.literal_eval(fl)
763 old_value = str(test_dict[k])
764 new_value = str(output[k])
772 fold = float(old_value)
773 fnew = float(new_value)
774 diff = abs(fold - fnew)
776 print(
"%s: test failed, old value: %s new value %s; "
777 "diff %f > %f" % (str(k), str(old_value),
778 str(new_value), diff,
779 tolerance), file=sys.stderr)
781 elif test_dict[k] != output[k]:
782 if len(old_value) < 50
and len(new_value) < 50:
783 print(
"%s: test failed, old value: %s new value %s"
784 % (str(k), old_value, new_value),
788 print(
"%s: test failed, omitting results (too long)"
789 % str(k), file=sys.stderr)
793 print(
"%s from old objects (file %s) not in new objects"
794 % (str(k), str(name)), file=sys.stderr)
798 def get_environment_variables(self):
800 return str(os.environ)
802 def get_versions_of_relevant_modules(self):
809 versions[
"ISD2_VERSION"] = IMP.isd2.get_module_version()
814 versions[
"ISD_EMXL_VERSION"] = IMP.isd_emxl.get_module_version()
819 def init_stat2(self, name, listofobjects, extralabels=None,
820 listofsummedobjects=
None):
827 if listofsummedobjects
is None:
828 listofsummedobjects = []
829 if extralabels
is None:
831 flstat = open(name,
'w')
833 stat2_keywords = {
"STAT2HEADER":
"STAT2HEADER"}
834 stat2_keywords.update(
835 {
"STAT2HEADER_ENVIRON": str(self.get_environment_variables())})
836 stat2_keywords.update(
837 {
"STAT2HEADER_IMP_VERSIONS":
838 str(self.get_versions_of_relevant_modules())})
841 for obj
in listofobjects:
842 if "get_output" not in dir(obj):
844 "Output: object %s doesn't have get_output() method"
849 dfiltered = dict((k, v)
850 for k, v
in d.items()
if k[0] !=
"_")
851 output.update(dfiltered)
854 for obj
in listofsummedobjects:
856 if "get_output" not in dir(t):
858 "Output: object %s doesn't have get_output() method"
861 if "_TotalScore" not in t.get_output():
863 "Output: object %s doesn't have _TotalScore "
864 "entry to be summed" % str(t))
866 output.update({obj[1]: 0.0})
868 for k
in extralabels:
869 output.update({k: 0.0})
871 for n, k
in enumerate(output):
872 stat2_keywords.update({n: k})
873 stat2_inverse.update({k: n})
875 flstat.write(
"%s \n" % stat2_keywords)
877 self.dictionary_stats2[name] = (
883 def write_stat2(self, name, appendmode=True):
885 (listofobjects, stat2_inverse, listofsummedobjects,
886 extralabels) = self.dictionary_stats2[name]
889 for obj
in listofobjects:
890 od = obj.get_output()
891 dfiltered = dict((k, v)
for k, v
in od.items()
if k[0] !=
"_")
893 output.update({stat2_inverse[k]: od[k]})
896 for so
in listofsummedobjects:
900 partial_score += float(d[
"_TotalScore"])
901 output.update({stat2_inverse[so[1]]: str(partial_score)})
904 for k
in extralabels:
905 if k
in self.initoutput:
906 output.update({stat2_inverse[k]: self.initoutput[k]})
908 output.update({stat2_inverse[k]:
"None"})
915 flstat = open(name, writeflag)
916 flstat.write(
"%s \n" % output)
919 def write_stats2(self):
920 for stat
in self.dictionary_stats2.keys():
921 self.write_stat2(stat)
925 """Collect statistics from ProcessOutput.get_fields().
926 Counters of the total number of frames read, plus the models that
927 passed the various filters used in get_fields(), are provided."""
930 self.passed_get_every = 0
931 self.passed_filterout = 0
932 self.passed_filtertuple = 0
936 """A class for reading stat files (either rmf or ascii v1 and v2)"""
937 def __init__(self, filename):
938 self.filename = filename
943 if self.filename
is None:
944 raise ValueError(
"No file name provided. Use -h for help")
948 rh = RMF.open_rmf_file_read_only(self.filename)
950 cat = rh.get_category(
'stat')
951 rmf_klist = rh.get_keys(cat)
952 self.rmf_names_keys = dict([(rh.get_name(k), k)
957 f = open(self.filename,
"r")
960 for line
in f.readlines():
961 d = ast.literal_eval(line)
962 self.klist = list(d.keys())
964 if "STAT2HEADER" in self.klist:
967 if "STAT2HEADER" in str(k):
973 for k
in sorted(stat2_dict.items(),
974 key=operator.itemgetter(1))]
976 for k
in sorted(stat2_dict.items(),
977 key=operator.itemgetter(1))]
978 self.invstat2_dict = {}
980 self.invstat2_dict.update({stat2_dict[k]: k})
983 "statfile v1 is deprecated. "
984 "Please convert to statfile v2.\n")
993 return sorted(self.rmf_names_keys.keys())
997 def show_keys(self, ncolumns=2, truncate=65):
998 IMP.pmi.tools.print_multicolumn(self.get_keys(), ncolumns, truncate)
1000 def get_fields(self, fields, filtertuple=None, filterout=None, get_every=1,
1003 Get the desired field names, and return a dictionary.
1004 Namely, "fields" are the queried keys in the stat file
1005 (eg. ["Total_Score",...])
1006 The returned data structure is a dictionary, where each key is
1007 a field and the value is the time series (ie, frame ordered series)
1008 of that field (ie, {"Total_Score":[Score_0,Score_1,Score_2,,...],....})
1010 @param fields (list of strings) queried keys in the stat file
1011 (eg. "Total_Score"....)
1012 @param filterout specify if you want to "grep" out something from
1013 the file, so that it is faster
1014 @param filtertuple a tuple that contains
1015 ("TheKeyToBeFiltered",relationship,value)
1016 where relationship = "<", "==", or ">"
1017 @param get_every only read every Nth line from the file
1018 @param statistics if provided, accumulate statistics in an
1019 OutputStatistics object
1022 if statistics
is None:
1025 for field
in fields:
1030 rh = RMF.open_rmf_file_read_only(self.filename)
1031 nframes = rh.get_number_of_frames()
1032 for i
in range(nframes):
1033 statistics.total += 1
1035 statistics.passed_get_every += 1
1036 statistics.passed_filterout += 1
1037 rh.set_current_frame(RMF.FrameID(i))
1038 if filtertuple
is not None:
1039 keytobefiltered = filtertuple[0]
1040 relationship = filtertuple[1]
1041 value = filtertuple[2]
1042 datavalue = rh.get_root_node().get_value(
1043 self.rmf_names_keys[keytobefiltered])
1044 if self.isfiltered(datavalue, relationship, value):
1047 statistics.passed_filtertuple += 1
1048 for field
in fields:
1049 outdict[field].append(rh.get_root_node().get_value(
1050 self.rmf_names_keys[field]))
1053 f = open(self.filename,
"r")
1056 for line
in f.readlines():
1057 statistics.total += 1
1058 if filterout
is not None:
1059 if filterout
in line:
1061 statistics.passed_filterout += 1
1064 if line_number % get_every != 0:
1065 if line_number == 1
and self.isstat2:
1066 statistics.total -= 1
1067 statistics.passed_filterout -= 1
1069 statistics.passed_get_every += 1
1071 d = ast.literal_eval(line)
1073 print(
"# Warning: skipped line number " + str(line_number)
1074 +
" not a valid line")
1079 if filtertuple
is not None:
1080 keytobefiltered = filtertuple[0]
1081 relationship = filtertuple[1]
1082 value = filtertuple[2]
1083 datavalue = d[keytobefiltered]
1084 if self.isfiltered(datavalue, relationship, value):
1087 statistics.passed_filtertuple += 1
1088 [outdict[field].append(d[field])
for field
in fields]
1091 if line_number == 1:
1092 statistics.total -= 1
1093 statistics.passed_filterout -= 1
1094 statistics.passed_get_every -= 1
1097 if filtertuple
is not None:
1098 keytobefiltered = filtertuple[0]
1099 relationship = filtertuple[1]
1100 value = filtertuple[2]
1101 datavalue = d[self.invstat2_dict[keytobefiltered]]
1102 if self.isfiltered(datavalue, relationship, value):
1105 statistics.passed_filtertuple += 1
1106 [outdict[field].append(d[self.invstat2_dict[field]])
1107 for field
in fields]
1113 def isfiltered(self, datavalue, relationship, refvalue):
1116 _ = float(datavalue)
1118 raise ValueError(
"ProcessOutput.filter: datavalue cannot be "
1119 "converted into a float")
1121 if relationship ==
"<":
1122 if float(datavalue) >= refvalue:
1124 if relationship ==
">":
1125 if float(datavalue) <= refvalue:
1127 if relationship ==
"==":
1128 if float(datavalue) != refvalue:
1134 """ class to allow more advanced handling of RMF files.
1135 It is both a container and a IMP.atom.Hierarchy.
1136 - it is iterable (while loading the corresponding frame)
1137 - Item brackets [] load the corresponding frame
1138 - slice create an iterator
1139 - can relink to another RMF file
1143 @param model: the IMP.Model()
1144 @param rmf_file_name: str, path of the rmf file
1148 self.rh_ref = RMF.open_rmf_file_read_only(rmf_file_name)
1150 raise TypeError(
"Wrong rmf file name or type: %s"
1151 % str(rmf_file_name))
1154 self.root_hier_ref = hs[0]
1155 IMP.atom.Hierarchy.__init__(self, self.root_hier_ref)
1157 self.ColorHierarchy =
None
1161 Link to another RMF file
1163 self.rh_ref = RMF.open_rmf_file_read_only(rmf_file_name)
1165 if self.ColorHierarchy:
1166 self.ColorHierarchy.method()
1167 RMFHierarchyHandler.set_frame(self, 0)
1169 def set_frame(self, index):
1173 print(
"skipping frame %s:%d\n" % (self.current_rmf, index))
1177 return self.rh_ref.get_number_of_frames()
1179 def __getitem__(self, int_slice_adaptor):
1180 if isinstance(int_slice_adaptor, int):
1181 self.set_frame(int_slice_adaptor)
1182 return int_slice_adaptor
1183 elif isinstance(int_slice_adaptor, slice):
1184 return self.__iter__(int_slice_adaptor)
1186 raise TypeError(
"Unknown Type")
1189 return self.get_number_of_frames()
1191 def __iter__(self, slice_key=None):
1192 if slice_key
is None:
1193 for nframe
in range(len(self)):
1196 for nframe
in list(range(len(self)))[slice_key]:
1200 class CacheHierarchyCoordinates(object):
1201 def __init__(self, StatHierarchyHandler):
1208 self.current_index =
None
1209 self.rmfh = StatHierarchyHandler
1211 self.model = self.rmfh.get_model()
1216 self.nrms.append(nrm)
1219 self.xyzs.append(fb)
1221 def do_store(self, index):
1222 self.rb_trans[index] = {}
1223 self.nrm_coors[index] = {}
1224 self.xyz_coors[index] = {}
1226 self.rb_trans[index][rb] = rb.get_reference_frame()
1227 for nrm
in self.nrms:
1228 self.nrm_coors[index][nrm] = nrm.get_internal_coordinates()
1229 for xyz
in self.xyzs:
1230 self.xyz_coors[index][xyz] = xyz.get_coordinates()
1231 self.current_index = index
1233 def do_update(self, index):
1234 if self.current_index != index:
1236 rb.set_reference_frame(self.rb_trans[index][rb])
1237 for nrm
in self.nrms:
1238 nrm.set_internal_coordinates(self.nrm_coors[index][nrm])
1239 for xyz
in self.xyzs:
1240 xyz.set_coordinates(self.xyz_coors[index][xyz])
1241 self.current_index = index
1245 return len(self.rb_trans.keys())
1247 def __getitem__(self, index):
1248 if isinstance(index, int):
1249 return index
in self.rb_trans.keys()
1251 raise TypeError(
"Unknown Type")
1254 return self.get_number_of_frames()
1258 """ class to link stat files to several rmf files """
1259 def __init__(self, model=None, stat_file=None,
1260 number_best_scoring_models=
None, score_key=
None,
1261 StatHierarchyHandler=
None, cache=
None):
1264 @param model: IMP.Model()
1265 @param stat_file: either 1) a list or 2) a single stat file names
1266 (either rmfs or ascii, or pickled data or pickled cluster),
1267 3) a dictionary containing an rmf/ascii
1268 stat file name as key and a list of frames as values
1269 @param number_best_scoring_models:
1270 @param StatHierarchyHandler: copy constructor input object
1271 @param cache: cache coordinates and rigid body transformations.
1274 if StatHierarchyHandler
is not None:
1278 self.model = StatHierarchyHandler.model
1279 self.data = StatHierarchyHandler.data
1280 self.number_best_scoring_models = \
1281 StatHierarchyHandler.number_best_scoring_models
1282 self.is_setup =
True
1283 self.current_rmf = StatHierarchyHandler.current_rmf
1284 self.current_frame =
None
1285 self.current_index =
None
1286 self.score_threshold = StatHierarchyHandler.score_threshold
1287 self.score_key = StatHierarchyHandler.score_key
1288 self.cache = StatHierarchyHandler.cache
1289 RMFHierarchyHandler.__init__(self, self.model,
1292 self.cache = CacheHierarchyCoordinates(self)
1301 self.number_best_scoring_models = number_best_scoring_models
1304 if score_key
is None:
1305 self.score_key =
"Total_Score"
1307 self.score_key = score_key
1308 self.is_setup =
None
1309 self.current_rmf =
None
1310 self.current_frame =
None
1311 self.current_index =
None
1312 self.score_threshold =
None
1314 if isinstance(stat_file, str):
1315 self.add_stat_file(stat_file)
1316 elif isinstance(stat_file, list):
1318 self.add_stat_file(f)
1320 def add_stat_file(self, stat_file):
1322 '''check that it is not a pickle file with saved data
1323 from a previous calculation'''
1324 self.load_data(stat_file)
1326 if self.number_best_scoring_models:
1327 scores = self.get_scores()
1328 max_score = sorted(scores)[
1329 0:min(len(self), self.number_best_scoring_models)][-1]
1330 self.do_filter_by_score(max_score)
1332 except pickle.UnpicklingError:
1333 '''alternatively read the ascii stat files'''
1335 scores, rmf_files, rmf_frame_indexes, features = \
1336 self.get_info_from_stat_file(stat_file,
1337 self.score_threshold)
1338 except (KeyError, SyntaxError):
1343 rh = RMF.open_rmf_file_read_only(stat_file)
1344 nframes = rh.get_number_of_frames()
1345 scores = [0.0]*nframes
1346 rmf_files = [stat_file]*nframes
1347 rmf_frame_indexes = range(nframes)
1352 if len(set(rmf_files)) > 1:
1353 raise (
"Multiple RMF files found")
1356 print(
"StatHierarchyHandler: Error: Trying to set none as "
1357 "rmf_file (probably empty stat file), aborting")
1360 for n, index
in enumerate(rmf_frame_indexes):
1361 featn_dict = dict([(k, features[k][n])
for k
in features])
1363 stat_file, rmf_files[n], index, scores[n], featn_dict))
1365 if self.number_best_scoring_models:
1366 scores = self.get_scores()
1367 max_score = sorted(scores)[
1368 0:min(len(self), self.number_best_scoring_models)][-1]
1369 self.do_filter_by_score(max_score)
1371 if not self.is_setup:
1372 RMFHierarchyHandler.__init__(
1373 self, self.model, self.get_rmf_names()[0])
1375 self.cache = CacheHierarchyCoordinates(self)
1378 self.is_setup =
True
1379 self.current_rmf = self.get_rmf_names()[0]
1383 def save_data(self, filename='data.pkl'):
1384 with open(filename,
'wb')
as fl:
1385 pickle.dump(self.data, fl)
1387 def load_data(self, filename='data.pkl'):
1388 with open(filename,
'rb')
as fl:
1389 data_structure = pickle.load(fl)
1391 if not isinstance(data_structure, list):
1393 "%filename should contain a list of IMP.pmi.output.DataEntry "
1394 "or IMP.pmi.output.Cluster" % filename)
1397 for item
in data_structure):
1398 self.data = data_structure
1400 for item
in data_structure):
1402 for cluster
in data_structure:
1403 nmodels += len(cluster)
1404 self.data = [
None]*nmodels
1405 for cluster
in data_structure:
1406 for n, data
in enumerate(cluster):
1407 index = cluster.members[n]
1408 self.data[index] = data
1411 "%filename should contain a list of IMP.pmi.output.DataEntry "
1412 "or IMP.pmi.output.Cluster" % filename)
1414 def set_frame(self, index):
1415 if self.cache
is not None and self.cache[index]:
1416 self.cache.do_update(index)
1418 nm = self.data[index].rmf_name
1419 fidx = self.data[index].rmf_index
1420 if nm != self.current_rmf:
1422 self.current_rmf = nm
1423 self.current_frame = -1
1424 if fidx != self.current_frame:
1425 RMFHierarchyHandler.set_frame(self, fidx)
1426 self.current_frame = fidx
1427 if self.cache
is not None:
1428 self.cache.do_store(index)
1430 self.current_index = index
1432 def __getitem__(self, int_slice_adaptor):
1433 if isinstance(int_slice_adaptor, int):
1434 self.set_frame(int_slice_adaptor)
1435 return self.data[int_slice_adaptor]
1436 elif isinstance(int_slice_adaptor, slice):
1437 return self.__iter__(int_slice_adaptor)
1439 raise TypeError(
"Unknown Type")
1442 return len(self.data)
1444 def __iter__(self, slice_key=None):
1445 if slice_key
is None:
1446 for i
in range(len(self)):
1449 for i
in range(len(self))[slice_key]:
1452 def do_filter_by_score(self, maximum_score):
1453 self.data = [d
for d
in self.data
if d.score <= maximum_score]
1455 def get_scores(self):
1456 return [d.score
for d
in self.data]
1458 def get_feature_series(self, feature_name):
1459 return [d.features[feature_name]
for d
in self.data]
1461 def get_feature_names(self):
1462 return self.data[0].features.keys()
1464 def get_rmf_names(self):
1465 return [d.rmf_name
for d
in self.data]
1467 def get_stat_files_names(self):
1468 return [d.stat_file
for d
in self.data]
1470 def get_rmf_indexes(self):
1471 return [d.rmf_index
for d
in self.data]
1473 def get_info_from_stat_file(self, stat_file, score_threshold=None):
1477 [stat_file], score_key=self.score_key, feature_keys=fs,
1478 rmf_file_key=
"rmf_file", rmf_file_frame_key=
"rmf_frame_index",
1479 prefiltervalue=score_threshold, get_every=1)
1481 scores = [float(y)
for y
in models[2]]
1482 rmf_files = models[0]
1483 rmf_frame_indexes = models[1]
1484 features = models[3]
1485 return scores, rmf_files, rmf_frame_indexes, features
1490 A class to store data associated to a model
1492 def __init__(self, stat_file=None, rmf_name=None, rmf_index=None,
1493 score=
None, features=
None):
1494 self.rmf_name = rmf_name
1495 self.rmf_index = rmf_index
1497 self.features = features
1498 self.stat_file = stat_file
1501 s =
"IMP.pmi.output.DataEntry\n"
1502 s +=
"---- stat file %s \n" % (self.stat_file)
1503 s +=
"---- rmf file %s \n" % (self.rmf_name)
1504 s +=
"---- rmf index %s \n" % (str(self.rmf_index))
1505 s +=
"---- score %s \n" % (str(self.score))
1506 s +=
"---- number of features %s \n" % (str(len(self.features.keys())))
1512 A container for models organized into clusters
1514 def __init__(self, cid=None):
1515 self.cluster_id = cid
1517 self.precision =
None
1518 self.center_index =
None
1519 self.members_data = {}
1521 def add_member(self, index, data=None):
1522 self.members.append(index)
1523 self.members_data[index] = data
1524 self.average_score = self.compute_score()
1526 def compute_score(self):
1528 score = sum([d.score
for d
in self])/len(self)
1529 except AttributeError:
1534 s =
"IMP.pmi.output.Cluster\n"
1535 s +=
"---- cluster_id %s \n" % str(self.cluster_id)
1536 s +=
"---- precision %s \n" % str(self.precision)
1537 s +=
"---- average score %s \n" % str(self.average_score)
1538 s +=
"---- number of members %s \n" % str(len(self.members))
1539 s +=
"---- center index %s \n" % str(self.center_index)
1542 def __getitem__(self, int_slice_adaptor):
1543 if isinstance(int_slice_adaptor, int):
1544 index = self.members[int_slice_adaptor]
1545 return self.members_data[index]
1546 elif isinstance(int_slice_adaptor, slice):
1547 return self.__iter__(int_slice_adaptor)
1549 raise TypeError(
"Unknown Type")
1552 return len(self.members)
1554 def __iter__(self, slice_key=None):
1555 if slice_key
is None:
1556 for i
in range(len(self)):
1559 for i
in range(len(self))[slice_key]:
1562 def __add__(self, other):
1563 self.members += other.members
1564 self.members_data.update(other.members_data)
1565 self.average_score = self.compute_score()
1566 self.precision =
None
1567 self.center_index =
None
1571 def plot_clusters_populations(clusters):
1574 for cluster
in clusters:
1575 indexes.append(cluster.cluster_id)
1576 populations.append(len(cluster))
1578 import matplotlib.pyplot
as plt
1579 fig, ax = plt.subplots()
1580 ax.bar(indexes, populations, 0.5, color=
'r')
1581 ax.set_ylabel('Population')
1582 ax.set_xlabel((
'Cluster index'))
1586 def plot_clusters_precisions(clusters):
1589 for cluster
in clusters:
1590 indexes.append(cluster.cluster_id)
1592 prec = cluster.precision
1593 print(cluster.cluster_id, prec)
1596 precisions.append(prec)
1598 import matplotlib.pyplot
as plt
1599 fig, ax = plt.subplots()
1600 ax.bar(indexes, precisions, 0.5, color=
'r')
1601 ax.set_ylabel('Precision [A]')
1602 ax.set_xlabel((
'Cluster index'))
1606 def plot_clusters_scores(clusters):
1609 for cluster
in clusters:
1610 indexes.append(cluster.cluster_id)
1612 for data
in cluster:
1613 values[-1].append(data.score)
1616 valuename=
"Scores", positionname=
"Cluster index",
1617 xlabels=
None, scale_plot_length=1.0)
1620 class CrossLinkIdentifierDatabase(object):
1624 def check_key(self, key):
1625 if key
not in self.clidb:
1626 self.clidb[key] = {}
1628 def set_unique_id(self, key, value):
1630 self.clidb[key][
"XLUniqueID"] = str(value)
1632 def set_protein1(self, key, value):
1634 self.clidb[key][
"Protein1"] = str(value)
1636 def set_protein2(self, key, value):
1638 self.clidb[key][
"Protein2"] = str(value)
1640 def set_residue1(self, key, value):
1642 self.clidb[key][
"Residue1"] = int(value)
1644 def set_residue2(self, key, value):
1646 self.clidb[key][
"Residue2"] = int(value)
1648 def set_idscore(self, key, value):
1650 self.clidb[key][
"IDScore"] = float(value)
1652 def set_state(self, key, value):
1654 self.clidb[key][
"State"] = int(value)
1656 def set_sigma1(self, key, value):
1658 self.clidb[key][
"Sigma1"] = str(value)
1660 def set_sigma2(self, key, value):
1662 self.clidb[key][
"Sigma2"] = str(value)
1664 def set_psi(self, key, value):
1666 self.clidb[key][
"Psi"] = str(value)
1668 def get_unique_id(self, key):
1669 return self.clidb[key][
"XLUniqueID"]
1671 def get_protein1(self, key):
1672 return self.clidb[key][
"Protein1"]
1674 def get_protein2(self, key):
1675 return self.clidb[key][
"Protein2"]
1677 def get_residue1(self, key):
1678 return self.clidb[key][
"Residue1"]
1680 def get_residue2(self, key):
1681 return self.clidb[key][
"Residue2"]
1683 def get_idscore(self, key):
1684 return self.clidb[key][
"IDScore"]
1686 def get_state(self, key):
1687 return self.clidb[key][
"State"]
1689 def get_sigma1(self, key):
1690 return self.clidb[key][
"Sigma1"]
1692 def get_sigma2(self, key):
1693 return self.clidb[key][
"Sigma2"]
1695 def get_psi(self, key):
1696 return self.clidb[key][
"Psi"]
1698 def set_float_feature(self, key, value, feature_name):
1700 self.clidb[key][feature_name] = float(value)
1702 def set_int_feature(self, key, value, feature_name):
1704 self.clidb[key][feature_name] = int(value)
1706 def set_string_feature(self, key, value, feature_name):
1708 self.clidb[key][feature_name] = str(value)
1710 def get_feature(self, key, feature_name):
1711 return self.clidb[key][feature_name]
1713 def write(self, filename):
1714 with open(filename,
'wb')
as handle:
1715 pickle.dump(self.clidb, handle)
1717 def load(self, filename):
1718 with open(filename,
'rb')
as handle:
1719 self.clidb = pickle.load(handle)
1723 """Plot the given fields and save a figure as `output`.
1724 The fields generally are extracted from a stat file
1725 using ProcessOutput.get_fields()."""
1726 import matplotlib
as mpl
1728 import matplotlib.pyplot
as plt
1730 plt.rc(
'lines', linewidth=4)
1731 fig, axs = plt.subplots(nrows=len(fields))
1732 fig.set_size_inches(10.5, 5.5 * len(fields))
1737 if framemin
is None:
1739 if framemax
is None:
1740 framemax = len(fields[key])
1741 x = list(range(framemin, framemax))
1742 y = [float(y)
for y
in fields[key][framemin:framemax]]
1745 axs[n].set_title(key, size=
"xx-large")
1746 axs[n].tick_params(labelsize=18, pad=10)
1749 axs.set_title(key, size=
"xx-large")
1750 axs.tick_params(labelsize=18, pad=10)
1754 plt.subplots_adjust(hspace=0.3)
1759 colors=
None, format=
"png", reference_xline=
None,
1760 yplotrange=
None, xplotrange=
None, normalized=
True,
1762 '''Plot a list of histograms from a value list.
1763 @param name the name of the plot
1764 @param values_lists the list of list of values eg: [[...],[...],[...]]
1765 @param valuename the y-label
1766 @param bins the number of bins
1767 @param colors If None, will use rainbow. Else will use specific list
1768 @param format output format
1769 @param reference_xline plot a reference line parallel to the y-axis
1770 @param yplotrange the range for the y-axis
1771 @param normalized whether the histogram is normalized or not
1772 @param leg_names names for the legend
1775 import matplotlib
as mpl
1777 import matplotlib.pyplot
as plt
1778 import matplotlib.cm
as cm
1779 plt.figure(figsize=(18.0, 9.0))
1782 colors = cm.rainbow(np.linspace(0, 1, len(values_lists)))
1783 for nv, values
in enumerate(values_lists):
1785 if leg_names
is not None:
1786 label = leg_names[nv]
1791 [float(y)
for y
in values], bins=bins, color=col,
1792 density=normalized, histtype=
'step', lw=4, label=label)
1793 except AttributeError:
1795 [float(y)
for y
in values], bins=bins, color=col,
1796 normed=normalized, histtype=
'step', lw=4, label=label)
1799 plt.tick_params(labelsize=12, pad=10)
1800 if valuename
is None:
1801 plt.xlabel(name, size=
"xx-large")
1803 plt.xlabel(valuename, size=
"xx-large")
1804 plt.ylabel(
"Frequency", size=
"xx-large")
1806 if yplotrange
is not None:
1808 if xplotrange
is not None:
1809 plt.xlim(xplotrange)
1813 if reference_xline
is not None:
1820 plt.savefig(name +
"." + format, dpi=150, transparent=
True)
1824 valuename=
"None", positionname=
"None",
1825 xlabels=
None, scale_plot_length=1.0):
1827 Plot time series as boxplots.
1828 fields is a list of time series, positions are the x-values
1829 valuename is the y-label, positionname is the x-label
1832 import matplotlib
as mpl
1834 import matplotlib.pyplot
as plt
1837 fig = plt.figure(figsize=(float(len(positions))*scale_plot_length, 5.0))
1838 fig.canvas.manager.set_window_title(name)
1840 ax1 = fig.add_subplot(111)
1842 plt.subplots_adjust(left=0.1, right=0.990, top=0.95, bottom=0.4)
1844 bps.append(plt.boxplot(values, notch=0, sym=
'', vert=1,
1845 whis=1.5, positions=positions))
1847 plt.setp(bps[-1][
'boxes'], color=
'black', lw=1.5)
1848 plt.setp(bps[-1][
'whiskers'], color=
'black', ls=
":", lw=1.5)
1850 if frequencies
is not None:
1851 for n, v
in enumerate(values):
1852 plist = [positions[n]]*len(v)
1853 ax1.plot(plist, v,
'gx', alpha=0.7, markersize=7)
1856 if xlabels
is not None:
1857 ax1.set_xticklabels(xlabels)
1858 plt.xticks(rotation=90)
1859 plt.xlabel(positionname)
1860 plt.ylabel(valuename)
1862 plt.savefig(name +
".pdf", dpi=150)
1866 def plot_xy_data(x, y, title=None, out_fn=None, display=True,
1867 set_plot_yaxis_range=
None, xlabel=
None, ylabel=
None):
1868 import matplotlib
as mpl
1870 import matplotlib.pyplot
as plt
1871 plt.rc(
'lines', linewidth=2)
1873 fig, ax = plt.subplots(nrows=1)
1874 fig.set_size_inches(8, 4.5)
1875 if title
is not None:
1876 fig.canvas.manager.set_window_title(title)
1878 ax.plot(x, y, color=
'r')
1879 if set_plot_yaxis_range
is not None:
1880 x1, x2, y1, y2 = plt.axis()
1881 y1 = set_plot_yaxis_range[0]
1882 y2 = set_plot_yaxis_range[1]
1883 plt.axis((x1, x2, y1, y2))
1884 if title
is not None:
1886 if xlabel
is not None:
1887 ax.set_xlabel(xlabel)
1888 if ylabel
is not None:
1889 ax.set_ylabel(ylabel)
1890 if out_fn
is not None:
1891 plt.savefig(out_fn +
".pdf")
1897 def plot_scatter_xy_data(x, y, labelx="None", labely="None",
1898 xmin=
None, xmax=
None, ymin=
None, ymax=
None,
1899 savefile=
False, filename=
"None.eps", alpha=0.75):
1901 import matplotlib
as mpl
1903 import matplotlib.pyplot
as plt
1904 from matplotlib
import rc
1905 rc(
'font', **{
'family':
'sans-serif',
'sans-serif': [
'Helvetica']})
1907 fig, axs = plt.subplots(1)
1911 axs0.set_xlabel(labelx, size=
"xx-large")
1912 axs0.set_ylabel(labely, size=
"xx-large")
1913 axs0.tick_params(labelsize=18, pad=10)
1917 plot2.append(axs0.plot(x, y,
'o', color=
'k', lw=2, ms=0.1, alpha=alpha,
1927 fig.set_size_inches(8.0, 8.0)
1928 fig.subplots_adjust(left=0.161, right=0.850, top=0.95, bottom=0.11)
1929 if (ymin
is not None)
and (ymax
is not None):
1930 axs0.set_ylim(ymin, ymax)
1931 if (xmin
is not None)
and (xmax
is not None):
1932 axs0.set_xlim(xmin, xmax)
1935 fig.savefig(filename, dpi=300)
1938 def get_graph_from_hierarchy(hier):
1942 (graph, depth, depth_dict) = recursive_graph(
1943 hier, graph, depth, depth_dict)
1946 node_labels_dict = {}
1947 for key
in depth_dict:
1948 if depth_dict[key] < 3:
1949 node_labels_dict[key] = key
1951 node_labels_dict[key] =
""
1952 draw_graph(graph, labels_dict=node_labels_dict)
1955 def recursive_graph(hier, graph, depth, depth_dict):
1958 index = str(hier.get_particle().
get_index())
1959 name1 = nameh +
"|#" + index
1960 depth_dict[name1] = depth
1964 if len(children) == 1
or children
is None:
1966 return (graph, depth, depth_dict)
1970 (graph, depth, depth_dict) = recursive_graph(
1971 c, graph, depth, depth_dict)
1973 index = str(c.get_particle().
get_index())
1974 namec = nameh +
"|#" + index
1975 graph.append((name1, namec))
1978 return (graph, depth, depth_dict)
1981 def draw_graph(graph, labels_dict=None, graph_layout='spring',
1982 node_size=5, node_color=
None, node_alpha=0.3,
1983 node_text_size=11, fixed=
None, pos=
None,
1984 edge_color=
'blue', edge_alpha=0.3, edge_thickness=1,
1986 validation_edges=
None,
1987 text_font=
'sans-serif',
1990 import matplotlib
as mpl
1992 import networkx
as nx
1993 import matplotlib.pyplot
as plt
1994 from math
import sqrt, pi
2000 if isinstance(edge_thickness, list):
2001 for edge, weight
in zip(graph, edge_thickness):
2002 G.add_edge(edge[0], edge[1], weight=weight)
2005 G.add_edge(edge[0], edge[1])
2007 if node_color
is None:
2008 node_color_rgb = (0, 0, 0)
2009 node_color_hex =
"000000"
2014 for node
in G.nodes():
2015 cctuple = cc.rgb(node_color[node])
2016 tmpcolor_rgb.append((cctuple[0]/255,
2019 tmpcolor_hex.append(node_color[node])
2020 node_color_rgb = tmpcolor_rgb
2021 node_color_hex = tmpcolor_hex
2024 if isinstance(node_size, dict):
2026 for node
in G.nodes():
2027 size = sqrt(node_size[node])/pi*10.0
2028 tmpsize.append(size)
2031 for n, node
in enumerate(G.nodes()):
2032 color = node_color_hex[n]
2034 nx.set_node_attributes(
2036 {node: {
'type':
'ellipse',
'w': size,
'h': size,
2037 'fill':
'#' + color,
'label': node}})
2038 nx.set_node_attributes(
2040 {node: {
'type':
'text',
'text': node,
'color':
'#000000',
2041 'visible':
'true'}})
2043 for edge
in G.edges():
2044 nx.set_edge_attributes(
2046 {edge: {
'width': 1,
'fill':
'#000000'}})
2048 for ve
in validation_edges:
2050 if (ve[0], ve[1])
in G.edges():
2051 print(
"found forward")
2052 nx.set_edge_attributes(
2054 {ve: {
'width': 1,
'fill':
'#00FF00'}})
2055 elif (ve[1], ve[0])
in G.edges():
2056 print(
"found backward")
2057 nx.set_edge_attributes(
2059 {(ve[1], ve[0]): {
'width': 1,
'fill':
'#00FF00'}})
2061 G.add_edge(ve[0], ve[1])
2063 nx.set_edge_attributes(
2065 {ve: {
'width': 1,
'fill':
'#FF0000'}})
2069 if graph_layout ==
'spring':
2071 graph_pos = nx.spring_layout(G, k=1.0/8.0, fixed=fixed, pos=pos)
2072 elif graph_layout ==
'spectral':
2073 graph_pos = nx.spectral_layout(G)
2074 elif graph_layout ==
'random':
2075 graph_pos = nx.random_layout(G)
2077 graph_pos = nx.shell_layout(G)
2080 nx.draw_networkx_nodes(G, graph_pos, node_size=node_size,
2081 alpha=node_alpha, node_color=node_color_rgb,
2083 nx.draw_networkx_edges(G, graph_pos, width=edge_thickness,
2084 alpha=edge_alpha, edge_color=edge_color)
2085 nx.draw_networkx_labels(
2086 G, graph_pos, labels=labels_dict, font_size=node_text_size,
2087 font_family=text_font)
2089 plt.savefig(out_filename)
2090 nx.write_gml(G,
'out.gml')
2098 from ipyD3
import d3object
2099 from IPython.display
import display
2101 d3 = d3object(width=800,
2106 title=
'Example table with d3js',
2107 desc=
'An example table created created with d3js with '
2108 'data generated with Python.')
2109 data = [[1277.0, 654.0, 288.0, 1976.0, 3281.0, 3089.0, 10336.0, 4650.0,
2110 4441.0, 4670.0, 944.0, 110.0],
2111 [1318.0, 664.0, 418.0, 1952.0, 3581.0, 4574.0, 11457.0, 6139.0,
2112 7078.0, 6561.0, 2354.0, 710.0],
2113 [1783.0, 774.0, 564.0, 1470.0, 3571.0, 3103.0, 9392.0, 5532.0,
2114 5661.0, 4991.0, 2032.0, 680.0],
2115 [1301.0, 604.0, 286.0, 2152.0, 3282.0, 3369.0, 10490.0, 5406.0,
2116 4727.0, 3428.0, 1559.0, 620.0],
2117 [1537.0, 1714.0, 724.0, 4824.0, 5551.0, 8096.0, 16589.0, 13650.0,
2118 9552.0, 13709.0, 2460.0, 720.0],
2119 [5691.0, 2995.0, 1680.0, 11741.0, 16232.0, 14731.0, 43522.0,
2120 32794.0, 26634.0, 31400.0, 7350.0, 3010.0],
2121 [1650.0, 2096.0, 60.0, 50.0, 1180.0, 5602.0, 15728.0, 6874.0,
2122 5115.0, 3510.0, 1390.0, 170.0],
2123 [72.0, 60.0, 60.0, 10.0, 120.0, 172.0, 1092.0, 675.0, 408.0,
2124 360.0, 156.0, 100.0]]
2125 data = [list(i)
for i
in zip(*data)]
2126 sRows = [[
'January',
2138 sColumns = [[
'Prod {0}'.format(i)
for i
in range(1, 9)],
2139 [
None,
'',
None,
None,
'Group 1',
None,
None,
'Group 2']]
2140 d3.addSimpleTable(data,
2141 fontSizeCells=[12, ],
2144 sRowsMargins=[5, 50, 0],
2145 sColsMargins=[5, 20, 10],
2148 addOutsideBorders=-1,
2152 html = d3.render(mode=[
'html',
'show'])
static bool get_is_setup(const IMP::ParticleAdaptor &p)
A container for models organized into clusters.
A class for reading stat files (either rmf or ascii v1 and v2)
atom::Hierarchies create_hierarchies(RMF::FileConstHandle fh, Model *m)
RMF::FrameID save_frame(RMF::FileHandle file, std::string name="")
Save the current state of the linked objects as a new RMF frame.
static bool get_is_setup(const IMP::ParticleAdaptor &p)
def plot_field_histogram
Plot a list of histograms from a value list.
def plot_fields_box_plots
Plot time series as boxplots.
Utility classes and functions for reading and storing PMI files.
def get_best_models
Given a list of stat files, read them all and find the best models.
A class to store data associated to a model.
void handle_use_deprecated(std::string message)
Break in this method in gdb to find deprecated uses at runtime.
std::string get_module_version()
Return the version of this module, as a string.
void write_pdb(const Selection &mhd, TextOutput out, unsigned int model=1)
Collect statistics from ProcessOutput.get_fields().
static bool get_is_setup(const IMP::ParticleAdaptor &p)
def get_fields
Get the desired field names, and return a dictionary.
Warning related to handling of structures.
static bool get_is_setup(Model *m, ParticleIndex pi)
def link_to_rmf
Link to another RMF file.
std::string get_molecule_name_and_copy(atom::Hierarchy h)
Walk up a PMI2 hierarchy/representations and get the "molname.copynum".
The standard decorator for manipulating molecular structures.
Ints get_index(const ParticlesTemp &particles, const Subset &subset, const Subsets &excluded)
def init_pdb
Init PDB Writing.
int get_number_of_frames(const ::npctransport_proto::Assignment &config, double time_step)
A decorator for a particle representing an atom.
Base class for capturing a modeling protocol.
void load_frame(RMF::FileConstHandle file, RMF::FrameID frame)
Load the given RMF frame into the state of the linked objects.
A decorator for a particle with x,y,z coordinates.
void add_hierarchies(RMF::NodeHandle fh, const atom::Hierarchies &hs)
Class for easy writing of PDBs, RMFs, and stat files.
void add_geometries(RMF::NodeHandle parent, const display::GeometriesTemp &r)
Add geometries to a given parent node.
void add_restraints(RMF::NodeHandle fh, const Restraints &hs)
A decorator for a particle that is part of a rigid body but not rigid.
Display a segment connecting a pair of particles.
A decorator for a residue.
Basic functionality that is expected to be used by a wide variety of IMP users.
def get_pdb_names
Get a list of all PDB files being output by this instance.
def get_prot_name_from_particle
Get the protein name from the particle.
class to link stat files to several rmf files
class to allow more advanced handling of RMF files.
void link_hierarchies(RMF::FileConstHandle fh, const atom::Hierarchies &hs)
def plot_fields
Plot the given fields and save a figure as output.
void add_geometry(RMF::FileHandle file, display::Geometry *r)
Add a single geometry to the file.
Store info for a chain of a protein.
Python classes to represent, score, sample and analyze models.
def init_pdb_best_scoring
Prepare for writing best-scoring PDBs (or mmCIFs) for a sampling run.
Functionality for loading, creating, manipulating and scoring atomic structures.
Select hierarchy particles identified by the biological name.
def init_rmf
Initialize an RMF file.
static bool get_is_setup(const IMP::ParticleAdaptor &p)
std::string get_module_version()
Return the version of this module, as a string.
A decorator for a particle with x,y,z coordinates and a radius.