1 """@namespace IMP.pmi.output
2 Classes for writing output files and processing them.
5 from __future__
import print_function, division
20 import cPickle
as pickle
24 class _ChainIDs(object):
25 """Map indices to multi-character chain IDs.
26 We label the first 26 chains A-Z, then we move to two-letter
27 chain IDs: AA through AZ, then BA through BZ, through to ZZ.
28 This continues with longer chain IDs."""
29 def __getitem__(self, ind):
30 chars = string.ascii_uppercase
34 ids.append(chars[ind % lc])
36 ids.append(chars[ind])
37 return "".join(reversed(ids))
41 """Base class for capturing a modeling protocol.
42 Unlike simple output of model coordinates, a complete
43 protocol includes the input data used, details on the restraints,
44 sampling, and clustering, as well as output models.
45 Use via IMP.pmi.topology.System.add_protocol_output().
47 @see IMP.pmi.mmcif.ProtocolOutput for a concrete subclass that outputs
54 if isinstance(elt, (tuple, list)):
55 for elt2
in _flatten(elt):
61 """Class for easy writing of PDBs, RMFs, and stat files
63 @note Model should be updated prior to writing outputs.
65 def __init__(self, ascii=True,atomistic=False):
66 self.dictionary_pdbs = {}
67 self.dictionary_rmfs = {}
68 self.dictionary_stats = {}
69 self.dictionary_stats2 = {}
70 self.best_score_list =
None
71 self.nbestscoring =
None
73 self.replica_exchange =
False
78 self.chainids =
"ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789"
80 self.multi_chainids = _ChainIDs()
82 self.particle_infos_for_pdb = {}
83 self.atomistic=atomistic
87 """Get a list of all PDB files being output by this instance"""
88 return list(self.dictionary_pdbs.keys())
90 def get_rmf_names(self):
91 return list(self.dictionary_rmfs.keys())
93 def get_stat_names(self):
94 return list(self.dictionary_stats.keys())
96 def _init_dictchain(self, name, prot, multichar_chain=False):
97 self.dictchain[name] = {}
103 self.atomistic =
True
104 for n,mol
in enumerate(IMP.atom.get_by_type(prot,IMP.atom.MOLECULE_TYPE)):
108 chainids = self.multi_chainids
if multichar_chain
else self.chainids
109 for n, i
in enumerate(self.dictionary_pdbs[name].get_children()):
110 self.dictchain[name][i.get_name()] = chainids[n]
114 @param name The PDB filename
115 @param prot The hierarchy to write to this pdb file
116 @note if the PDB name is 'System' then will use Selection to get molecules
118 flpdb = open(name,
'w')
120 self.dictionary_pdbs[name] = prot
121 self._init_dictchain(name, prot)
123 def write_psf(self,filename,name):
124 flpsf=open(filename,
'w')
125 flpsf.write(
"PSF CMAP CHEQ"+
"\n")
126 index_residue_pair_list={}
127 (particle_infos_for_pdb, geometric_center)=self.get_particle_infos_for_pdb_writing(name)
128 nparticles=len(particle_infos_for_pdb)
129 flpsf.write(str(nparticles)+
" !NATOM"+
"\n")
130 for n,p
in enumerate(particle_infos_for_pdb):
135 flpsf.write(
'{0:8d}{1:1s}{2:4s}{3:1s}{4:4s}{5:1s}{6:4s}{7:1s}{8:4s}{9:1s}{10:4s}{11:14.6f}{12:14.6f}{13:8d}{14:14.6f}{15:14.6f}'.format(atom_index,
" ",chain,
" ",str(resid),
" ",
'"'+residue_type.get_string()+
'"',
" ",
"C",
" ",
"C",1.0,0.0,0,0.0,0.0))
138 if chain
not in index_residue_pair_list:
139 index_residue_pair_list[chain]=[(atom_index,resid)]
141 index_residue_pair_list[chain].append((atom_index,resid))
146 for chain
in sorted(index_residue_pair_list.keys()):
148 ls=index_residue_pair_list[chain]
150 ls=sorted(ls, key=
lambda tup: tup[1])
152 indexes=[x[0]
for x
in ls]
155 nbonds=len(indexes_pairs)
156 flpsf.write(str(nbonds)+
" !NBOND: bonds"+
"\n")
159 for i
in range(0,len(indexes_pairs),4):
160 for bond
in indexes_pairs[i:i+4]:
161 flpsf.write(
'{0:8d}{1:8d}'.format(*bond))
164 del particle_infos_for_pdb
169 translate_to_geometric_center=
False,
170 write_all_residues_per_bead=
False):
172 flpdb = open(name,
'a')
174 flpdb = open(name,
'w')
176 (particle_infos_for_pdb,
177 geometric_center) = self.get_particle_infos_for_pdb_writing(name)
179 if not translate_to_geometric_center:
180 geometric_center = (0, 0, 0)
182 for n,tupl
in enumerate(particle_infos_for_pdb):
183 (xyz, atom_type, residue_type,
184 chain_id, residue_index, all_indexes, radius) = tupl
185 if atom_type
is None:
186 atom_type = IMP.atom.AT_CA
187 if ( (write_all_residues_per_bead)
and (all_indexes
is not None) ):
188 for residue_number
in all_indexes:
189 flpdb.write(IMP.atom.get_pdb_string((xyz[0] - geometric_center[0],
190 xyz[1] - geometric_center[1],
191 xyz[2] - geometric_center[2]),
192 n+1, atom_type, residue_type,
193 chain_id, residue_number,
' ',1.00,radius))
195 flpdb.write(IMP.atom.get_pdb_string((xyz[0] - geometric_center[0],
196 xyz[1] - geometric_center[1],
197 xyz[2] - geometric_center[2]),
198 n+1, atom_type, residue_type,
199 chain_id, residue_index,
' ',1.00,radius))
200 flpdb.write(
"ENDMDL\n")
203 del particle_infos_for_pdb
206 """Get the protein name from the particle.
207 This is done by traversing the hierarchy."""
212 p, self.dictchain[name])
214 def get_particle_infos_for_pdb_writing(self, name):
224 particle_infos_for_pdb = []
226 geometric_center = [0, 0, 0]
232 ps =
IMP.atom.Selection(self.dictionary_pdbs[name],resolution=0).get_selected_particles()
236 for n, p
in enumerate(ps):
239 if protname
not in resindexes_dict:
240 resindexes_dict[protname] = []
244 rt = residue.get_residue_type()
245 resind = residue.get_index()
249 geometric_center[0] += xyz[0]
250 geometric_center[1] += xyz[1]
251 geometric_center[2] += xyz[2]
253 particle_infos_for_pdb.append((xyz,
254 atomtype, rt, self.dictchain[name][protname], resind,
None, radius))
255 resindexes_dict[protname].append(resind)
260 resind = residue.get_index()
263 if resind
in resindexes_dict[protname]:
266 resindexes_dict[protname].append(resind)
267 rt = residue.get_residue_type()
270 geometric_center[0] += xyz[0]
271 geometric_center[1] += xyz[1]
272 geometric_center[2] += xyz[2]
274 particle_infos_for_pdb.append((xyz,
None,
275 rt, self.dictchain[name][protname], resind,
None, radius))
279 resind = resindexes[len(resindexes) // 2]
280 if resind
in resindexes_dict[protname]:
283 resindexes_dict[protname].append(resind)
287 geometric_center[0] += xyz[0]
288 geometric_center[1] += xyz[1]
289 geometric_center[2] += xyz[2]
291 particle_infos_for_pdb.append((xyz,
None,
292 rt, self.dictchain[name][protname], resind, resindexes, radius))
298 if len(resindexes) > 0:
299 resind = resindexes[len(resindexes) // 2]
302 geometric_center[0] += xyz[0]
303 geometric_center[1] += xyz[1]
304 geometric_center[2] += xyz[2]
306 particle_infos_for_pdb.append((xyz,
None,
307 rt, self.dictchain[name][protname], resind, resindexes, radius))
310 geometric_center = (geometric_center[0] / atom_count,
311 geometric_center[1] / atom_count,
312 geometric_center[2] / atom_count)
316 particle_infos_for_pdb = sorted(particle_infos_for_pdb,
317 key=
lambda x: (len(x[3]), x[3], x[4]))
319 return (particle_infos_for_pdb, geometric_center)
322 def write_pdbs(self, appendmode=True):
323 for pdb
in self.dictionary_pdbs.keys():
324 self.write_pdb(pdb, appendmode)
326 def init_pdb_best_scoring(self,
330 replica_exchange=
False):
334 self.suffixes.append(suffix)
335 self.replica_exchange = replica_exchange
336 if not self.replica_exchange:
340 self.best_score_list = []
344 self.best_score_file_name =
"best.scores.rex.py"
345 self.best_score_list = []
346 with open(self.best_score_file_name,
"w")
as best_score_file:
347 best_score_file.write(
348 "self.best_score_list=" + str(self.best_score_list) +
"\n")
350 self.nbestscoring = nbestscoring
351 for i
in range(self.nbestscoring):
352 name = suffix +
"." + str(i) +
".pdb"
353 flpdb = open(name,
'w')
355 self.dictionary_pdbs[name] = prot
356 self._init_dictchain(name, prot)
358 def write_pdb_best_scoring(self, score):
359 if self.nbestscoring
is None:
360 print(
"Output.write_pdb_best_scoring: init_pdb_best_scoring not run")
363 if self.replica_exchange:
365 with open(self.best_score_file_name)
as fh:
368 if len(self.best_score_list) < self.nbestscoring:
369 self.best_score_list.append(score)
370 self.best_score_list.sort()
371 index = self.best_score_list.index(score)
372 for suffix
in self.suffixes:
373 for i
in range(len(self.best_score_list) - 2, index - 1, -1):
374 oldname = suffix +
"." + str(i) +
".pdb"
375 newname = suffix +
"." + str(i + 1) +
".pdb"
377 if os.path.exists(newname):
379 os.rename(oldname, newname)
380 filetoadd = suffix +
"." + str(index) +
".pdb"
381 self.write_pdb(filetoadd, appendmode=
False)
384 if score < self.best_score_list[-1]:
385 self.best_score_list.append(score)
386 self.best_score_list.sort()
387 self.best_score_list.pop(-1)
388 index = self.best_score_list.index(score)
389 for suffix
in self.suffixes:
390 for i
in range(len(self.best_score_list) - 1, index - 1, -1):
391 oldname = suffix +
"." + str(i) +
".pdb"
392 newname = suffix +
"." + str(i + 1) +
".pdb"
393 os.rename(oldname, newname)
394 filenametoremove = suffix + \
395 "." + str(self.nbestscoring) +
".pdb"
396 os.remove(filenametoremove)
397 filetoadd = suffix +
"." + str(index) +
".pdb"
398 self.write_pdb(filetoadd, appendmode=
False)
400 if self.replica_exchange:
402 with open(self.best_score_file_name,
"w")
as best_score_file:
403 best_score_file.write(
404 "self.best_score_list=" + str(self.best_score_list) +
'\n')
406 def init_rmf(self, name, hierarchies, rs=None, geometries=None, listofobjects=None):
408 Initialize an RMF file
410 @param name the name of the RMF file
411 @param hierarchies the hierarchies to be included (it is a list)
412 @param rs optional, the restraint sets (it is a list)
413 @param geometries optional, the geometries (it is a list)
414 @param listofobjects optional, the list of objects for the stat (it is a list)
416 rh = RMF.create_rmf_file(name)
419 outputkey_rmfkey=
None
423 if geometries
is not None:
425 if listofobjects
is not None:
426 cat = rh.get_category(
"stat")
428 for l
in listofobjects:
429 if not "get_output" in dir(l):
430 raise ValueError(
"Output: object %s doesn't have get_output() method" % str(l))
431 output=l.get_output()
432 for outputkey
in output:
433 rmftag=RMF.string_tag
434 if isinstance(output[outputkey], float):
435 rmftag = RMF.float_tag
436 elif isinstance(output[outputkey], int):
438 elif isinstance(output[outputkey], str):
439 rmftag = RMF.string_tag
441 rmftag = RMF.string_tag
442 rmfkey=rh.get_key(cat, outputkey, rmftag)
443 outputkey_rmfkey[outputkey]=rmfkey
444 outputkey_rmfkey[
"rmf_file"]=rh.get_key(cat,
"rmf_file", RMF.string_tag)
445 outputkey_rmfkey[
"rmf_frame_index"]=rh.get_key(cat,
"rmf_frame_index", RMF.int_tag)
447 self.dictionary_rmfs[name] = (rh,cat,outputkey_rmfkey,listofobjects)
449 def add_restraints_to_rmf(self, name, objectlist):
450 for o
in _flatten(objectlist):
452 rs = o.get_restraint_for_rmf()
454 rs = o.get_restraint()
456 self.dictionary_rmfs[name][0],
459 def add_geometries_to_rmf(self, name, objectlist):
461 geos = o.get_geometries()
464 def add_particle_pair_from_restraints_to_rmf(self, name, objectlist):
467 pps = o.get_particle_pairs()
470 self.dictionary_rmfs[name][0],
473 def write_rmf(self, name):
475 if self.dictionary_rmfs[name][1]
is not None:
476 cat=self.dictionary_rmfs[name][1]
477 outputkey_rmfkey=self.dictionary_rmfs[name][2]
478 listofobjects=self.dictionary_rmfs[name][3]
479 for l
in listofobjects:
480 output=l.get_output()
481 for outputkey
in output:
482 rmfkey=outputkey_rmfkey[outputkey]
484 self.dictionary_rmfs[name][0].get_root_node().set_value(rmfkey,output[outputkey])
485 except NotImplementedError:
487 rmfkey = outputkey_rmfkey[
"rmf_file"]
488 self.dictionary_rmfs[name][0].get_root_node().set_value(rmfkey, name)
489 rmfkey = outputkey_rmfkey[
"rmf_frame_index"]
491 self.dictionary_rmfs[name][0].get_root_node().set_value(rmfkey, nframes-1)
492 self.dictionary_rmfs[name][0].flush()
494 def close_rmf(self, name):
495 rh = self.dictionary_rmfs[name][0]
496 del self.dictionary_rmfs[name]
499 def write_rmfs(self):
500 for rmfinfo
in self.dictionary_rmfs.keys():
501 self.write_rmf(rmfinfo[0])
503 def init_stat(self, name, listofobjects):
505 flstat = open(name,
'w')
508 flstat = open(name,
'wb')
512 for l
in listofobjects:
513 if not "get_output" in dir(l):
514 raise ValueError(
"Output: object %s doesn't have get_output() method" % str(l))
515 self.dictionary_stats[name] = listofobjects
517 def set_output_entry(self, key, value):
518 self.initoutput.update({key: value})
520 def write_stat(self, name, appendmode=True):
521 output = self.initoutput
522 for obj
in self.dictionary_stats[name]:
525 dfiltered = dict((k, v)
for k, v
in d.items()
if k[0] !=
"_")
526 output.update(dfiltered)
534 flstat = open(name, writeflag)
535 flstat.write(
"%s \n" % output)
538 flstat = open(name, writeflag +
'b')
539 cPickle.dump(output, flstat, 2)
542 def write_stats(self):
543 for stat
in self.dictionary_stats.keys():
544 self.write_stat(stat)
546 def get_stat(self, name):
548 for obj
in self.dictionary_stats[name]:
549 output.update(obj.get_output())
552 def write_test(self, name, listofobjects):
559 flstat = open(name,
'w')
560 output = self.initoutput
561 for l
in listofobjects:
562 if not "get_test_output" in dir(l)
and not "get_output" in dir(l):
563 raise ValueError(
"Output: object %s doesn't have get_output() or get_test_output() method" % str(l))
564 self.dictionary_stats[name] = listofobjects
566 for obj
in self.dictionary_stats[name]:
568 d = obj.get_test_output()
572 dfiltered = dict((k, v)
for k, v
in d.items()
if k[0] !=
"_")
573 output.update(dfiltered)
577 flstat.write(
"%s \n" % output)
580 def test(self, name, listofobjects, tolerance=1e-5):
581 output = self.initoutput
582 for l
in listofobjects:
583 if not "get_test_output" in dir(l)
and not "get_output" in dir(l):
584 raise ValueError(
"Output: object %s doesn't have get_output() or get_test_output() method" % str(l))
585 for obj
in listofobjects:
587 output.update(obj.get_test_output())
589 output.update(obj.get_output())
594 flstat = open(name,
'r')
598 test_dict = ast.literal_eval(l)
601 old_value = str(test_dict[k])
602 new_value = str(output[k])
610 fold = float(old_value)
611 fnew = float(new_value)
612 diff = abs(fold - fnew)
614 print(
"%s: test failed, old value: %s new value %s; "
615 "diff %f > %f" % (str(k), str(old_value),
616 str(new_value), diff,
617 tolerance), file=sys.stderr)
619 elif test_dict[k] != output[k]:
620 if len(old_value) < 50
and len(new_value) < 50:
621 print(
"%s: test failed, old value: %s new value %s"
622 % (str(k), old_value, new_value), file=sys.stderr)
625 print(
"%s: test failed, omitting results (too long)"
626 % str(k), file=sys.stderr)
630 print(
"%s from old objects (file %s) not in new objects"
631 % (str(k), str(name)), file=sys.stderr)
635 def get_environment_variables(self):
637 return str(os.environ)
639 def get_versions_of_relevant_modules(self):
646 versions[
"ISD2_VERSION"] = IMP.isd2.get_module_version()
647 except (ImportError):
651 versions[
"ISD_EMXL_VERSION"] = IMP.isd_emxl.get_module_version()
652 except (ImportError):
662 listofsummedobjects=
None):
668 if listofsummedobjects
is None:
669 listofsummedobjects = []
670 if extralabels
is None:
672 flstat = open(name,
'w')
674 stat2_keywords = {
"STAT2HEADER":
"STAT2HEADER"}
675 stat2_keywords.update(
676 {
"STAT2HEADER_ENVIRON": str(self.get_environment_variables())})
677 stat2_keywords.update(
678 {
"STAT2HEADER_IMP_VERSIONS": str(self.get_versions_of_relevant_modules())})
681 for l
in listofobjects:
682 if not "get_output" in dir(l):
683 raise ValueError(
"Output: object %s doesn't have get_output() method" % str(l))
687 dfiltered = dict((k, v)
688 for k, v
in d.items()
if k[0] !=
"_")
689 output.update(dfiltered)
692 for l
in listofsummedobjects:
694 if not "get_output" in dir(t):
695 raise ValueError(
"Output: object %s doesn't have get_output() method" % str(t))
697 if "_TotalScore" not in t.get_output():
698 raise ValueError(
"Output: object %s doesn't have _TotalScore entry to be summed" % str(t))
700 output.update({l[1]: 0.0})
702 for k
in extralabels:
703 output.update({k: 0.0})
705 for n, k
in enumerate(output):
706 stat2_keywords.update({n: k})
707 stat2_inverse.update({k: n})
709 flstat.write(
"%s \n" % stat2_keywords)
711 self.dictionary_stats2[name] = (
717 def write_stat2(self, name, appendmode=True):
719 (listofobjects, stat2_inverse, listofsummedobjects,
720 extralabels) = self.dictionary_stats2[name]
723 for obj
in listofobjects:
724 od = obj.get_output()
725 dfiltered = dict((k, v)
for k, v
in od.items()
if k[0] !=
"_")
727 output.update({stat2_inverse[k]: od[k]})
730 for l
in listofsummedobjects:
734 partial_score += float(d[
"_TotalScore"])
735 output.update({stat2_inverse[l[1]]: str(partial_score)})
738 for k
in extralabels:
739 if k
in self.initoutput:
740 output.update({stat2_inverse[k]: self.initoutput[k]})
742 output.update({stat2_inverse[k]:
"None"})
749 flstat = open(name, writeflag)
750 flstat.write(
"%s \n" % output)
753 def write_stats2(self):
754 for stat
in self.dictionary_stats2.keys():
755 self.write_stat2(stat)
759 """Collect statistics from ProcessOutput.get_fields().
760 Counters of the total number of frames read, plus the models that
761 passed the various filters used in get_fields(), are provided."""
764 self.passed_get_every = 0
765 self.passed_filterout = 0
766 self.passed_filtertuple = 0
770 """A class for reading stat files (either rmf or ascii v1 and v2)"""
771 def __init__(self, filename):
772 self.filename = filename
777 if self.filename
is None:
778 raise ValueError(
"No file name provided. Use -h for help")
782 rh = RMF.open_rmf_file_read_only(self.filename)
784 cat=rh.get_category(
'stat')
785 rmf_klist=rh.get_keys(cat)
786 self.rmf_names_keys=dict([(rh.get_name(k),k)
for k
in rmf_klist])
790 f = open(self.filename,
"r")
793 for line
in f.readlines():
794 d = ast.literal_eval(line)
795 self.klist = list(d.keys())
797 if "STAT2HEADER" in self.klist:
800 if "STAT2HEADER" in str(k):
806 for k
in sorted(stat2_dict.items(), key=operator.itemgetter(1))]
808 for k
in sorted(stat2_dict.items(), key=operator.itemgetter(1))]
809 self.invstat2_dict = {}
811 self.invstat2_dict.update({stat2_dict[k]: k})
814 "Please convert to statfile v2.\n")
824 return sorted(self.rmf_names_keys.keys())
828 def show_keys(self, ncolumns=2, truncate=65):
829 IMP.pmi.tools.print_multicolumn(self.get_keys(), ncolumns, truncate)
831 def get_fields(self, fields, filtertuple=None, filterout=None, get_every=1,
834 Get the desired field names, and return a dictionary.
835 Namely, "fields" are the queried keys in the stat file (eg. ["Total_Score",...])
836 The returned data structure is a dictionary, where each key is a field and the value
837 is the time series (ie, frame ordered series)
838 of that field (ie, {"Total_Score":[Score_0,Score_1,Score_2,Score_3,...],....} )
840 @param fields (list of strings) queried keys in the stat file (eg. "Total_Score"....)
841 @param filterout specify if you want to "grep" out something from
842 the file, so that it is faster
843 @param filtertuple a tuple that contains
844 ("TheKeyToBeFiltered",relationship,value)
845 where relationship = "<", "==", or ">"
846 @param get_every only read every Nth line from the file
847 @param statistics if provided, accumulate statistics in an
848 OutputStatistics object
851 if statistics
is None:
859 rh = RMF.open_rmf_file_read_only(self.filename)
860 nframes=rh.get_number_of_frames()
861 for i
in range(nframes):
862 statistics.total += 1
864 statistics.passed_get_every += 1
865 statistics.passed_filterout += 1
867 if not filtertuple
is None:
868 keytobefiltered = filtertuple[0]
869 relationship = filtertuple[1]
870 value = filtertuple[2]
871 datavalue=rh.get_root_node().get_value(self.rmf_names_keys[keytobefiltered])
872 if self.isfiltered(datavalue,relationship,value):
continue
874 statistics.passed_filtertuple += 1
876 outdict[field].append(rh.get_root_node().get_value(self.rmf_names_keys[field]))
879 f = open(self.filename,
"r")
882 for line
in f.readlines():
883 statistics.total += 1
884 if not filterout
is None:
885 if filterout
in line:
887 statistics.passed_filterout += 1
890 if line_number % get_every != 0:
891 if line_number == 1
and self.isstat2:
892 statistics.total -= 1
893 statistics.passed_filterout -= 1
895 statistics.passed_get_every += 1
899 d = ast.literal_eval(line)
901 print(
"# Warning: skipped line number " + str(line_number) +
" not a valid line")
906 if not filtertuple
is None:
907 keytobefiltered = filtertuple[0]
908 relationship = filtertuple[1]
909 value = filtertuple[2]
910 datavalue=d[keytobefiltered]
911 if self.isfiltered(datavalue, relationship, value):
continue
913 statistics.passed_filtertuple += 1
914 [outdict[field].append(d[field])
for field
in fields]
918 statistics.total -= 1
919 statistics.passed_filterout -= 1
920 statistics.passed_get_every -= 1
923 if not filtertuple
is None:
924 keytobefiltered = filtertuple[0]
925 relationship = filtertuple[1]
926 value = filtertuple[2]
927 datavalue=d[self.invstat2_dict[keytobefiltered]]
928 if self.isfiltered(datavalue, relationship, value):
continue
930 statistics.passed_filtertuple += 1
931 [outdict[field].append(d[self.invstat2_dict[field]])
for field
in fields]
937 def isfiltered(self,datavalue,relationship,refvalue):
940 fdatavalue=float(datavalue)
942 raise ValueError(
"ProcessOutput.filter: datavalue cannot be converted into a float")
944 if relationship ==
"<":
945 if float(datavalue) >= refvalue:
947 if relationship ==
">":
948 if float(datavalue) <= refvalue:
950 if relationship ==
"==":
951 if float(datavalue) != refvalue:
957 """ class to allow more advanced handling of RMF files.
958 It is both a container and a IMP.atom.Hierarchy.
959 - it is iterable (while loading the corresponding frame)
960 - Item brackets [] load the corresponding frame
961 - slice create an iterator
962 - can relink to another RMF file
966 @param model: the IMP.Model()
967 @param rmf_file_name: str, path of the rmf file
971 self.rh_ref = RMF.open_rmf_file_read_only(rmf_file_name)
973 raise TypeError(
"Wrong rmf file name or type: %s"% str(rmf_file_name))
976 self.root_hier_ref = hs[0]
977 IMP.atom.Hierarchy.__init__(self, self.root_hier_ref)
979 self.ColorHierarchy=
None
984 Link to another RMF file
986 self.rh_ref = RMF.open_rmf_file_read_only(rmf_file_name)
988 if self.ColorHierarchy:
989 self.ColorHierarchy.method()
990 RMFHierarchyHandler.set_frame(self,0)
992 def set_frame(self,index):
996 print(
"skipping frame %s:%d\n"%(self.current_rmf, index))
1000 return self.rh_ref.get_number_of_frames()
1002 def __getitem__(self,int_slice_adaptor):
1003 if isinstance(int_slice_adaptor, int):
1004 self.set_frame(int_slice_adaptor)
1005 return int_slice_adaptor
1006 elif isinstance(int_slice_adaptor, slice):
1007 return self.__iter__(int_slice_adaptor)
1009 raise TypeError(
"Unknown Type")
1012 return self.get_number_of_frames()
1014 def __iter__(self,slice_key=None):
1015 if slice_key
is None:
1016 for nframe
in range(len(self)):
1019 for nframe
in list(range(len(self)))[slice_key]:
1022 class CacheHierarchyCoordinates(object):
1023 def __init__(self,StatHierarchyHandler):
1030 self.current_index=
None
1031 self.rmfh=StatHierarchyHandler
1033 self.model=self.rmfh.get_model()
1038 self.nrms.append(nrm)
1041 self.xyzs.append(fb)
1043 def do_store(self,index):
1044 self.rb_trans[index]={}
1045 self.nrm_coors[index]={}
1046 self.xyz_coors[index]={}
1048 self.rb_trans[index][rb]=rb.get_reference_frame()
1049 for nrm
in self.nrms:
1050 self.nrm_coors[index][nrm]=nrm.get_internal_coordinates()
1051 for xyz
in self.xyzs:
1052 self.xyz_coors[index][xyz]=xyz.get_coordinates()
1053 self.current_index=index
1055 def do_update(self,index):
1056 if self.current_index!=index:
1058 rb.set_reference_frame(self.rb_trans[index][rb])
1059 for nrm
in self.nrms:
1060 nrm.set_internal_coordinates(self.nrm_coors[index][nrm])
1061 for xyz
in self.xyzs:
1062 xyz.set_coordinates(self.xyz_coors[index][xyz])
1063 self.current_index=index
1067 return len(self.rb_trans.keys())
1069 def __getitem__(self,index):
1070 if isinstance(index, int):
1071 return index
in self.rb_trans.keys()
1073 raise TypeError(
"Unknown Type")
1076 return self.get_number_of_frames()
1082 """ class to link stat files to several rmf files """
1083 def __init__(self,model=None,stat_file=None,number_best_scoring_models=None,score_key=None,StatHierarchyHandler=None,cache=None):
1086 @param model: IMP.Model()
1087 @param stat_file: either 1) a list or 2) a single stat file names (either rmfs or ascii, or pickled data or pickled cluster), 3) a dictionary containing an rmf/ascii
1088 stat file name as key and a list of frames as values
1089 @param number_best_scoring_models:
1090 @param StatHierarchyHandler: copy constructor input object
1091 @param cache: cache coordinates and rigid body transformations.
1094 if not StatHierarchyHandler
is None:
1097 self.model=StatHierarchyHandler.model
1098 self.data=StatHierarchyHandler.data
1099 self.number_best_scoring_models=StatHierarchyHandler.number_best_scoring_models
1101 self.current_rmf=StatHierarchyHandler.current_rmf
1102 self.current_frame=
None
1103 self.current_index=
None
1104 self.score_threshold=StatHierarchyHandler.score_threshold
1105 self.score_key=StatHierarchyHandler.score_key
1106 self.cache=StatHierarchyHandler.cache
1107 RMFHierarchyHandler.__init__(self, self.model,self.current_rmf)
1109 self.cache=CacheHierarchyCoordinates(self)
1118 self.number_best_scoring_models=number_best_scoring_models
1121 if score_key
is None:
1122 self.score_key=
"Total_Score"
1124 self.score_key=score_key
1126 self.current_rmf=
None
1127 self.current_frame=
None
1128 self.current_index=
None
1129 self.score_threshold=
None
1131 if isinstance(stat_file, str):
1132 self.add_stat_file(stat_file)
1133 elif isinstance(stat_file, list):
1135 self.add_stat_file(f)
1137 def add_stat_file(self,stat_file):
1139 '''check that it is not a pickle file with saved data from a previous calculation'''
1140 self.load_data(stat_file)
1142 if self.number_best_scoring_models:
1143 scores = self.get_scores()
1144 max_score = sorted(scores)[0:min(len(self), self.number_best_scoring_models)][-1]
1145 self.do_filter_by_score(max_score)
1147 except pickle.UnpicklingError:
1148 '''alternatively read the ascii stat files'''
1150 scores,rmf_files,rmf_frame_indexes,features = self.get_info_from_stat_file(stat_file, self.score_threshold)
1151 except (KeyError, SyntaxError):
1155 rh = RMF.open_rmf_file_read_only(stat_file)
1156 nframes = rh.get_number_of_frames()
1157 scores=[0.0]*nframes
1158 rmf_files=[stat_file]*nframes
1159 rmf_frame_indexes=range(nframes)
1165 if len(set(rmf_files)) > 1:
1166 raise (
"Multiple RMF files found")
1169 print(
"StatHierarchyHandler: Error: Trying to set none as rmf_file (probably empty stat file), aborting")
1172 for n,index
in enumerate(rmf_frame_indexes):
1173 featn_dict=dict([(k,features[k][n])
for k
in features])
1176 if self.number_best_scoring_models:
1177 scores=self.get_scores()
1178 max_score=sorted(scores)[0:min(len(self),self.number_best_scoring_models)][-1]
1179 self.do_filter_by_score(max_score)
1181 if not self.is_setup:
1182 RMFHierarchyHandler.__init__(self, self.model,self.get_rmf_names()[0])
1184 self.cache=CacheHierarchyCoordinates(self)
1188 self.current_rmf=self.get_rmf_names()[0]
1192 def save_data(self,filename='data.pkl'):
1193 with open(filename,
'wb')
as fl:
1194 pickle.dump(self.data, fl)
1196 def load_data(self,filename='data.pkl'):
1197 with open(filename,
'rb')
as fl:
1198 data_structure=pickle.load(fl)
1200 if not isinstance(data_structure, list):
1201 raise TypeError(
"%filename should contain a list of IMP.pmi.output.DataEntry or IMP.pmi.output.Cluster" % filename)
1204 self.data=data_structure
1207 for cluster
in data_structure:
1208 nmodels+=len(cluster)
1209 self.data=[
None]*nmodels
1210 for cluster
in data_structure:
1211 for n,data
in enumerate(cluster):
1212 index=cluster.members[n]
1213 self.data[index]=data
1215 raise TypeError(
"%filename should contain a list of IMP.pmi.output.DataEntry or IMP.pmi.output.Cluster" % filename)
1217 def set_frame(self,index):
1218 if self.cache
is not None and self.cache[index]:
1219 self.cache.do_update(index)
1221 nm=self.data[index].rmf_name
1222 fidx=self.data[index].rmf_index
1223 if nm != self.current_rmf:
1226 self.current_frame=-1
1227 if fidx!=self.current_frame:
1228 RMFHierarchyHandler.set_frame(self, fidx)
1229 self.current_frame=fidx
1230 if self.cache
is not None:
1231 self.cache.do_store(index)
1233 self.current_index = index
1235 def __getitem__(self,int_slice_adaptor):
1236 if isinstance(int_slice_adaptor, int):
1237 self.set_frame(int_slice_adaptor)
1238 return self.data[int_slice_adaptor]
1239 elif isinstance(int_slice_adaptor, slice):
1240 return self.__iter__(int_slice_adaptor)
1242 raise TypeError(
"Unknown Type")
1245 return len(self.data)
1247 def __iter__(self,slice_key=None):
1248 if slice_key
is None:
1249 for i
in range(len(self)):
1252 for i
in range(len(self))[slice_key]:
1255 def do_filter_by_score(self,maximum_score):
1256 self.data=[d
for d
in self.data
if d.score<=maximum_score]
1258 def get_scores(self):
1259 return [d.score
for d
in self.data]
1261 def get_feature_series(self,feature_name):
1262 return [d.features[feature_name]
for d
in self.data]
1264 def get_feature_names(self):
1265 return self.data[0].features.keys()
1267 def get_rmf_names(self):
1268 return [d.rmf_name
for d
in self.data]
1270 def get_stat_files_names(self):
1271 return [d.stat_file
for d
in self.data]
1273 def get_rmf_indexes(self):
1274 return [d.rmf_index
for d
in self.data]
1276 def get_info_from_stat_file(self, stat_file, score_threshold=None):
1280 score_key=self.score_key,
1282 rmf_file_key=
"rmf_file",
1283 rmf_file_frame_key=
"rmf_frame_index",
1284 prefiltervalue=score_threshold,
1289 scores = [float(y)
for y
in models[2]]
1290 rmf_files = models[0]
1291 rmf_frame_indexes = models[1]
1293 return scores, rmf_files, rmf_frame_indexes,features
1298 A class to store data associated to a model
1300 def __init__(self,stat_file=None,rmf_name=None,rmf_index=None,score=None,features=None):
1301 self.rmf_name=rmf_name
1302 self.rmf_index=rmf_index
1304 self.features=features
1305 self.stat_file=stat_file
1308 s=
"IMP.pmi.output.DataEntry\n"
1309 s+=
"---- stat file %s \n"%(self.stat_file)
1310 s+=
"---- rmf file %s \n"%(self.rmf_name)
1311 s+=
"---- rmf index %s \n"%(str(self.rmf_index))
1312 s+=
"---- score %s \n"%(str(self.score))
1313 s+=
"---- number of features %s \n"%(str(len(self.features.keys())))
1319 A container for models organized into clusters
1321 def __init__(self,cid=None):
1325 self.center_index=
None
1326 self.members_data={}
1328 def add_member(self,index,data=None):
1329 self.members.append(index)
1330 self.members_data[index]=data
1331 self.average_score=self.compute_score()
1333 def compute_score(self):
1335 score=sum([d.score
for d
in self])/len(self)
1336 except AttributeError:
1341 s=
"IMP.pmi.output.Cluster\n"
1342 s+=
"---- cluster_id %s \n"%str(self.cluster_id)
1343 s+=
"---- precision %s \n"%str(self.precision)
1344 s+=
"---- average score %s \n"%str(self.average_score)
1345 s+=
"---- number of members %s \n"%str(len(self.members))
1346 s+=
"---- center index %s \n"%str(self.center_index)
1349 def __getitem__(self,int_slice_adaptor):
1350 if isinstance(int_slice_adaptor, int):
1351 index=self.members[int_slice_adaptor]
1352 return self.members_data[index]
1353 elif isinstance(int_slice_adaptor, slice):
1354 return self.__iter__(int_slice_adaptor)
1356 raise TypeError(
"Unknown Type")
1359 return len(self.members)
1361 def __iter__(self,slice_key=None):
1362 if slice_key
is None:
1363 for i
in range(len(self)):
1366 for i
in range(len(self))[slice_key]:
1369 def __add__(self, other):
1370 self.members+=other.members
1371 self.members_data.update(other.members_data)
1372 self.average_score=self.compute_score()
1374 self.center_index=
None
1378 def plot_clusters_populations(clusters):
1381 for cluster
in clusters:
1382 indexes.append(cluster.cluster_id)
1383 populations.append(len(cluster))
1385 import matplotlib.pyplot
as plt
1386 fig, ax = plt.subplots()
1387 ax.bar(indexes, populations, 0.5, color=
'r') #, yerr=men_std)
1388 ax.set_ylabel('Population')
1389 ax.set_xlabel((
'Cluster index'))
1392 def plot_clusters_precisions(clusters):
1395 for cluster
in clusters:
1396 indexes.append(cluster.cluster_id)
1398 prec=cluster.precision
1399 print(cluster.cluster_id,prec)
1402 precisions.append(prec)
1404 import matplotlib.pyplot
as plt
1405 fig, ax = plt.subplots()
1406 ax.bar(indexes, precisions, 0.5, color=
'r') #, yerr=men_std)
1407 ax.set_ylabel('Precision [A]')
1408 ax.set_xlabel((
'Cluster index'))
1411 def plot_clusters_scores(clusters):
1414 for cluster
in clusters:
1415 indexes.append(cluster.cluster_id)
1417 for data
in cluster:
1418 values[-1].append(data.score)
1421 valuename=
"Scores", positionname=
"Cluster index", xlabels=
None,scale_plot_length=1.0)
1423 class CrossLinkIdentifierDatabase(object):
1427 def check_key(self,key):
1428 if key
not in self.clidb:
1431 def set_unique_id(self,key,value):
1433 self.clidb[key][
"XLUniqueID"]=str(value)
1435 def set_protein1(self,key,value):
1437 self.clidb[key][
"Protein1"]=str(value)
1439 def set_protein2(self,key,value):
1441 self.clidb[key][
"Protein2"]=str(value)
1443 def set_residue1(self,key,value):
1445 self.clidb[key][
"Residue1"]=int(value)
1447 def set_residue2(self,key,value):
1449 self.clidb[key][
"Residue2"]=int(value)
1451 def set_idscore(self,key,value):
1453 self.clidb[key][
"IDScore"]=float(value)
1455 def set_state(self,key,value):
1457 self.clidb[key][
"State"]=int(value)
1459 def set_sigma1(self,key,value):
1461 self.clidb[key][
"Sigma1"]=str(value)
1463 def set_sigma2(self,key,value):
1465 self.clidb[key][
"Sigma2"]=str(value)
1467 def set_psi(self,key,value):
1469 self.clidb[key][
"Psi"]=str(value)
1471 def get_unique_id(self,key):
1472 return self.clidb[key][
"XLUniqueID"]
1474 def get_protein1(self,key):
1475 return self.clidb[key][
"Protein1"]
1477 def get_protein2(self,key):
1478 return self.clidb[key][
"Protein2"]
1480 def get_residue1(self,key):
1481 return self.clidb[key][
"Residue1"]
1483 def get_residue2(self,key):
1484 return self.clidb[key][
"Residue2"]
1486 def get_idscore(self,key):
1487 return self.clidb[key][
"IDScore"]
1489 def get_state(self,key):
1490 return self.clidb[key][
"State"]
1492 def get_sigma1(self,key):
1493 return self.clidb[key][
"Sigma1"]
1495 def get_sigma2(self,key):
1496 return self.clidb[key][
"Sigma2"]
1498 def get_psi(self,key):
1499 return self.clidb[key][
"Psi"]
1501 def set_float_feature(self,key,value,feature_name):
1503 self.clidb[key][feature_name]=float(value)
1505 def set_int_feature(self,key,value,feature_name):
1507 self.clidb[key][feature_name]=int(value)
1509 def set_string_feature(self,key,value,feature_name):
1511 self.clidb[key][feature_name]=str(value)
1513 def get_feature(self,key,feature_name):
1514 return self.clidb[key][feature_name]
1516 def write(self,filename):
1517 with open(filename,
'wb')
as handle:
1518 pickle.dump(self.clidb,handle)
1520 def load(self,filename):
1521 with open(filename,
'rb')
as handle:
1522 self.clidb=pickle.load(handle)
1524 def plot_fields(fields, framemin=None, framemax=None):
1525 import matplotlib
as mpl
1527 import matplotlib.pyplot
as plt
1529 plt.rc(
'lines', linewidth=4)
1530 fig, axs = plt.subplots(nrows=len(fields))
1531 fig.set_size_inches(10.5, 5.5 * len(fields))
1532 plt.rc(
'axes', color_cycle=[
'r'])
1536 if framemin
is None:
1538 if framemax
is None:
1539 framemax = len(fields[key])
1540 x = list(range(framemin, framemax))
1541 y = [float(y)
for y
in fields[key][framemin:framemax]]
1544 axs[n].set_title(key, size=
"xx-large")
1545 axs[n].tick_params(labelsize=18, pad=10)
1548 axs.set_title(key, size=
"xx-large")
1549 axs.tick_params(labelsize=18, pad=10)
1553 plt.subplots_adjust(hspace=0.3)
1558 name, values_lists, valuename=
None, bins=40, colors=
None, format=
"png",
1559 reference_xline=
None, yplotrange=
None, xplotrange=
None,normalized=
True,
1562 '''Plot a list of histograms from a value list.
1563 @param name the name of the plot
1564 @param value_lists the list of list of values eg: [[...],[...],[...]]
1565 @param valuename the y-label
1566 @param bins the number of bins
1567 @param colors If None, will use rainbow. Else will use specific list
1568 @param format output format
1569 @param reference_xline plot a reference line parallel to the y-axis
1570 @param yplotrange the range for the y-axis
1571 @param normalized whether the histogram is normalized or not
1572 @param leg_names names for the legend
1575 import matplotlib
as mpl
1577 import matplotlib.pyplot
as plt
1578 import matplotlib.cm
as cm
1579 fig = plt.figure(figsize=(18.0, 9.0))
1582 colors = cm.rainbow(np.linspace(0, 1, len(values_lists)))
1583 for nv,values
in enumerate(values_lists):
1585 if leg_names
is not None:
1590 [float(y)
for y
in values],
1593 normed=normalized,histtype=
'step',lw=4,
1597 plt.tick_params(labelsize=12, pad=10)
1598 if valuename
is None:
1599 plt.xlabel(name, size=
"xx-large")
1601 plt.xlabel(valuename, size=
"xx-large")
1602 plt.ylabel(
"Frequency", size=
"xx-large")
1604 if not yplotrange
is None:
1606 if not xplotrange
is None:
1607 plt.xlim(xplotrange)
1611 if not reference_xline
is None:
1618 plt.savefig(name +
"." + format, dpi=150, transparent=
True)
1623 valuename=
"None", positionname=
"None", xlabels=
None,scale_plot_length=1.0):
1625 Plot time series as boxplots.
1626 fields is a list of time series, positions are the x-values
1627 valuename is the y-label, positionname is the x-label
1630 import matplotlib
as mpl
1632 import matplotlib.pyplot
as plt
1633 from matplotlib.patches
import Polygon
1636 fig = plt.figure(figsize=(float(len(positions))*scale_plot_length, 5.0))
1637 fig.canvas.set_window_title(name)
1639 ax1 = fig.add_subplot(111)
1641 plt.subplots_adjust(left=0.1, right=0.990, top=0.95, bottom=0.4)
1643 bps.append(plt.boxplot(values, notch=0, sym=
'', vert=1,
1644 whis=1.5, positions=positions))
1646 plt.setp(bps[-1][
'boxes'], color=
'black', lw=1.5)
1647 plt.setp(bps[-1][
'whiskers'], color=
'black', ls=
":", lw=1.5)
1649 if frequencies
is not None:
1650 for n,v
in enumerate(values):
1651 plist=[positions[n]]*len(v)
1652 ax1.plot(plist, v,
'gx', alpha=0.7, markersize=7)
1655 if not xlabels
is None:
1656 ax1.set_xticklabels(xlabels)
1657 plt.xticks(rotation=90)
1658 plt.xlabel(positionname)
1659 plt.ylabel(valuename)
1661 plt.savefig(name+
".pdf",dpi=150)
1665 def plot_xy_data(x,y,title=None,out_fn=None,display=True,set_plot_yaxis_range=None,
1666 xlabel=
None,ylabel=
None):
1667 import matplotlib
as mpl
1669 import matplotlib.pyplot
as plt
1670 plt.rc(
'lines', linewidth=2)
1672 fig, ax = plt.subplots(nrows=1)
1673 fig.set_size_inches(8,4.5)
1674 if title
is not None:
1675 fig.canvas.set_window_title(title)
1678 ax.plot(x,y,color=
'r')
1679 if set_plot_yaxis_range
is not None:
1680 x1,x2,y1,y2=plt.axis()
1681 y1=set_plot_yaxis_range[0]
1682 y2=set_plot_yaxis_range[1]
1683 plt.axis((x1,x2,y1,y2))
1684 if title
is not None:
1686 if xlabel
is not None:
1687 ax.set_xlabel(xlabel)
1688 if ylabel
is not None:
1689 ax.set_ylabel(ylabel)
1690 if out_fn
is not None:
1691 plt.savefig(out_fn+
".pdf")
1696 def plot_scatter_xy_data(x,y,labelx="None",labely="None",
1697 xmin=
None,xmax=
None,ymin=
None,ymax=
None,
1698 savefile=
False,filename=
"None.eps",alpha=0.75):
1700 import matplotlib
as mpl
1702 import matplotlib.pyplot
as plt
1704 from matplotlib
import rc
1706 rc(
'font',**{
'family':
'sans-serif',
'sans-serif':[
'Helvetica']})
1709 fig, axs = plt.subplots(1)
1713 axs0.set_xlabel(labelx, size=
"xx-large")
1714 axs0.set_ylabel(labely, size=
"xx-large")
1715 axs0.tick_params(labelsize=18, pad=10)
1719 plot2.append(axs0.plot(x, y,
'o', color=
'k',lw=2, ms=0.1, alpha=alpha, c=
"w"))
1728 fig.set_size_inches(8.0, 8.0)
1729 fig.subplots_adjust(left=0.161, right=0.850, top=0.95, bottom=0.11)
1730 if (
not ymin
is None)
and (
not ymax
is None):
1731 axs0.set_ylim(ymin,ymax)
1732 if (
not xmin
is None)
and (
not xmax
is None):
1733 axs0.set_xlim(xmin,xmax)
1737 fig.savefig(filename, dpi=300)
1740 def get_graph_from_hierarchy(hier):
1744 (graph, depth, depth_dict) = recursive_graph(
1745 hier, graph, depth, depth_dict)
1748 node_labels_dict = {}
1750 for key
in depth_dict:
1751 node_size_dict = 10 / depth_dict[key]
1752 if depth_dict[key] < 3:
1753 node_labels_dict[key] = key
1755 node_labels_dict[key] =
""
1756 draw_graph(graph, labels_dict=node_labels_dict)
1759 def recursive_graph(hier, graph, depth, depth_dict):
1762 index = str(hier.get_particle().
get_index())
1763 name1 = nameh +
"|#" + index
1764 depth_dict[name1] = depth
1768 if len(children) == 1
or children
is None:
1770 return (graph, depth, depth_dict)
1774 (graph, depth, depth_dict) = recursive_graph(
1775 c, graph, depth, depth_dict)
1777 index = str(c.get_particle().
get_index())
1778 namec = nameh +
"|#" + index
1779 graph.append((name1, namec))
1782 return (graph, depth, depth_dict)
1785 def draw_graph(graph, labels_dict=None, graph_layout='spring',
1786 node_size=5, node_color=
None, node_alpha=0.3,
1787 node_text_size=11, fixed=
None, pos=
None,
1788 edge_color=
'blue', edge_alpha=0.3, edge_thickness=1,
1790 validation_edges=
None,
1791 text_font=
'sans-serif',
1794 import matplotlib
as mpl
1796 import networkx
as nx
1797 import matplotlib.pyplot
as plt
1798 from math
import sqrt, pi
1804 if isinstance(edge_thickness, list):
1805 for edge,weight
in zip(graph,edge_thickness):
1806 G.add_edge(edge[0], edge[1], weight=weight)
1809 G.add_edge(edge[0], edge[1])
1811 if node_color==
None:
1812 node_color_rgb=(0,0,0)
1813 node_color_hex=
"000000"
1818 for node
in G.nodes():
1819 cctuple=cc.rgb(node_color[node])
1820 tmpcolor_rgb.append((cctuple[0]/255,cctuple[1]/255,cctuple[2]/255))
1821 tmpcolor_hex.append(node_color[node])
1822 node_color_rgb=tmpcolor_rgb
1823 node_color_hex=tmpcolor_hex
1826 if isinstance(node_size, dict):
1828 for node
in G.nodes():
1829 size=sqrt(node_size[node])/pi*10.0
1830 tmpsize.append(size)
1833 for n,node
in enumerate(G.nodes()):
1834 color=node_color_hex[n]
1836 nx.set_node_attributes(G,
"graphics", {node : {
'type':
'ellipse',
'w': size,
'h': size,
'fill':
'#'+color,
'label': node}})
1837 nx.set_node_attributes(G,
"LabelGraphics", {node : {
'type':
'text',
'text':node,
'color':
'#000000',
'visible':
'true'}})
1839 for edge
in G.edges():
1840 nx.set_edge_attributes(G,
"graphics", {edge : {
'width': 1,
'fill':
'#000000'}})
1842 for ve
in validation_edges:
1844 if (ve[0],ve[1])
in G.edges():
1845 print(
"found forward")
1846 nx.set_edge_attributes(G,
"graphics", {ve : {
'width': 1,
'fill':
'#00FF00'}})
1847 elif (ve[1],ve[0])
in G.edges():
1848 print(
"found backward")
1849 nx.set_edge_attributes(G,
"graphics", {(ve[1],ve[0]) : {
'width': 1,
'fill':
'#00FF00'}})
1851 G.add_edge(ve[0], ve[1])
1853 nx.set_edge_attributes(G,
"graphics", {ve : {
'width': 1,
'fill':
'#FF0000'}})
1857 if graph_layout ==
'spring':
1859 graph_pos = nx.spring_layout(G,k=1.0/8.0,fixed=fixed,pos=pos)
1860 elif graph_layout ==
'spectral':
1861 graph_pos = nx.spectral_layout(G)
1862 elif graph_layout ==
'random':
1863 graph_pos = nx.random_layout(G)
1865 graph_pos = nx.shell_layout(G)
1869 nx.draw_networkx_nodes(G, graph_pos, node_size=node_size,
1870 alpha=node_alpha, node_color=node_color_rgb,
1872 nx.draw_networkx_edges(G, graph_pos, width=edge_thickness,
1873 alpha=edge_alpha, edge_color=edge_color)
1874 nx.draw_networkx_labels(
1875 G, graph_pos, labels=labels_dict, font_size=node_text_size,
1876 font_family=text_font)
1878 plt.savefig(out_filename)
1879 nx.write_gml(G,
'out.gml')
1887 from ipyD3
import d3object
1888 from IPython.display
import display
1890 d3 = d3object(width=800,
1895 title=
'Example table with d3js',
1896 desc=
'An example table created created with d3js with data generated with Python.')
1982 [72.0, 60.0, 60.0, 10.0, 120.0, 172.0, 1092.0, 675.0, 408.0, 360.0, 156.0, 100.0]]
1983 data = [list(i)
for i
in zip(*data)]
1984 sRows = [[
'January',
1996 sColumns = [[
'Prod {0}'.format(i)
for i
in range(1, 9)],
1997 [
None,
'',
None,
None,
'Group 1',
None,
None,
'Group 2']]
1998 d3.addSimpleTable(data,
1999 fontSizeCells=[12, ],
2002 sRowsMargins=[5, 50, 0],
2003 sColsMargins=[5, 20, 10],
2006 addOutsideBorders=-1,
2010 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)
std::string get_module_version()
void write_pdb(const Selection &mhd, TextOutput out, unsigned int model=1)
Collect statistics from ProcessOutput.get_fields().
def get_fields
Get the desired field names, and return a dictionary.
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.
bool get_is_canonical(atom::Hierarchy h)
Walk up a PMI2 hierarchy/representations and check if the root is named System.
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)
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.
Functionality for loading, creating, manipulating and scoring atomic structures.
Hierarchies get_leaves(const Selection &h)
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()
A decorator for a particle with x,y,z coordinates and a radius.