def main():
args = parse_args()
trajectory = load_pdb(args.trajectory)
average = calc_average(trajectory, args.cutoff)
prody.writePDB(args.out_file, average)
def extract_heteroatoms_pdbs(pdb,
create_file=True,
ligand_chain="L",
get_ligand=False,
output_folder="."):
"""
From a pdb file, it extracts the chain L and checks if the structure has hydrogens. After that, the chain L is
written in a new PDB file which will have the following format: "{residue name}.pdb".
:param pdb: pdb file (with a ligand in the chain L).
:return: Writes a new pdb file "{residue name}.pdb" with the chain L isolated an returns the residue name (string).
"""
# Parse the complex file and isolate the ligand core and the fragment
ligand = complex_to_prody.pdb_parser_ligand(pdb, ligand_chain)
if ligand is None:
logger.critical(
"The ligand can not be found. Ensure that the ligand of {} is the chain {}"
.format(pdb, ligand_chain))
# Check if the ligand has H
complex_to_prody.check_protonation(ligand)
# Save the ligand in a PDB (the name of the file is the name of the residue)
ligand_name = ligand.getResnames()[0]
if create_file:
prody.writePDB(os.path.join(output_folder, ligand_name), ligand)
print(
"The ligand of {} has been extracted and saved in '{}.pdb'".format(
pdb, os.path.join(output_folder, ligand_name)))
if get_ligand is True:
return ligand
else:
return ligand_name
def getResidueSasa(atoms, return_atom_asa=False):
protein_atoms = atoms.protein.noh
assert protein_atoms.getHierView().numChains() == 1
pdb_file = tempfile.NamedTemporaryFile(prefix='tmp_naccess_', suffix='.pdb', delete=False)
pdb_filename = pdb_file.name
prody.writePDB(pdb_filename, protein_atoms)
pdb_file.close()
prefix, ext = os.path.splitext(pdb_filename)
#print prefix, ext
rsa_filename = '{}.rsa'.format(prefix)
asa_filename = '{}.asa'.format(prefix)
# run naccess on the temporary PDB file, with $CWD set to
# the PDB file's directory. This is because naccess dumps
# all the output files in $CWD, and they aren't necessary
# after parsing.
subprocess.Popen(
[NACCESS_BINARY, pdb_filename, ],
cwd=os.path.dirname(pdb_filename),
).communicate()
# read the resulting .rsa file into a table
rsa_table = read_naccess_rsa(rsa_filename)
if return_atom_asa:
atom_asa = read_naccess_asa(asa_filename)
assert len(atom_asa) == len(protein_atoms), '{} != {}'.format(len(atom_asa), len(protein_atoms))
return rsa_table, atom_asa
return rsa_table
def selection_to_pdb(selection):
"""
:param selection: prody selection
:return: writes a PDB file containing the selection, named using the residue name of the first atom of the selection.
"""
prody.writePDB(selection.getResnames()[0], selection)
return "{}.pdb".format(selection.getResnames()[0])
def prody_biomol(pdbname, **kwargs):
"""Generate biomolecule coordinates.
:arg pdb: PDB identifier or filename
:arg prefix: prefix for output files, default is :file:`_biomol`
:arg biomol: index of the biomolecule, by default all are generated"""
import prody
LOGGER = prody.LOGGER
prefix, biomol = kwargs.get('prefix', None), kwargs.get('biomol')
pdb, header = prody.parsePDB(pdbname, header=True)
if not prefix:
prefix = pdb.getTitle()
biomols = prody.buildBiomolecules(header, pdb, biomol=biomol)
if not isinstance(biomols, list):
biomols = [biomols]
for i, biomol in enumerate(biomols):
if isinstance(biomol, prody.Atomic):
outfn = '{0}_biomol_{1}.pdb'.format(prefix, i + 1)
LOGGER.info('Writing {0}'.format(outfn))
prody.writePDB(outfn, biomol)
elif isinstance(biomol, tuple):
for j, part in enumerate(biomol):
outfn = ('{0}_biomol_{1}_part_{2}.pdb'.format(
prefix, i + 1, j + 1))
LOGGER.info('Writing {0}'.format(outfn))
prody.writePDB(outfn, part)
def prody_biomol(opt):
"""Generate biomolecule coordinates based on command line arguments."""
import prody
LOGGER = prody.LOGGER
prefix, biomol = opt.prefix, opt.biomol
pdb, header = prody.parsePDB(opt.pdb, header=True)
if not prefix:
prefix = pdb.getTitle()
biomols = prody.buildBiomolecules(header, pdb, biomol=biomol)
if not isinstance(biomols, list):
biomols = [biomols]
for i, biomol in enumerate(biomols):
if isinstance(biomol, prody.Atomic):
outfn = '{0:s}_biomol_{1:d}.pdb'.format(prefix, i+1)
LOGGER.info('Writing {0:s}'.format(outfn))
prody.writePDB(outfn, biomol)
elif isinstance(biomol, tuple):
for j, part in enumerate(biomol):
outfn = ('{0:s}_biomol_{1:d}_part_{2:d}.pdb'
.format(prefix, i+1, j+1))
LOGGER.info('Writing {0:s}'.format(outfn))
prody.writePDB(outfn, part)
def prody_biomol(pdbname,**kwargs):
"""Generate biomolecule coordinates.
:arg pdb: :term:`PDB` identifier or filename
:arg prefix: prefix for output files, default is :file:`_biomol`
:arg biomol: index of the biomolecule, by default all are generated"""
import prody
LOGGER = prody.LOGGER
prefix, biomol = kwargs.get('prefix',None), kwargs.get('biomol')
pdb, header = prody.parsePDB(pdbname, header=True)
if not prefix:
prefix = pdb.getTitle()
biomols = prody.buildBiomolecules(header, pdb, biomol=biomol)
if not isinstance(biomols, list):
biomols = [biomols]
for i, biomol in enumerate(biomols):
if isinstance(biomol, prody.Atomic):
outfn = '{0:s}_biomol_{1:d}.pdb'.format(prefix, i+1)
LOGGER.info('Writing {0:s}'.format(outfn))
prody.writePDB(outfn, biomol)
elif isinstance(biomol, tuple):
for j, part in enumerate(biomol):
outfn = ('{0:s}_biomol_{1:d}_part_{2:d}.pdb'
.format(prefix, i+1, j+1))
LOGGER.info('Writing {0:s}'.format(outfn))
prody.writePDB(outfn, part)
def main():
in_file, trajectory, out_file, force_const = parse_args()
average = load_pdb(in_file)
trajectory = load_pdb(trajectory)
starting_model = get_closest_frame(trajectory, average)
minimized_protein = run_minimization(average, starting_model, force_const)
prody.writePDB(out_file, minimized_protein)
def extract_ligand(self, path):
"""
Writes the ligand into a file.
:param path: path to write the ligand.
:return:
"""
prody.writePDB(path, self.selection)
def clean_pdb(pdb_file, chain, write_out=True):
pdb = pdy.parsePDB(pdb_file)
p = pdb.select(f'stdaa and chain {chain}')
# Ensure that all resnums are above 0
res_nums = [
res_num for res_num in sorted(list(set(p.getResnums()))) if res_num > 0
]
# Remove Insertion Codes
clean_res_num = []
for res_num in res_nums:
if len(set(p.select(f'resnum {res_num}').getIcodes())) > 1:
clean_res_num.append(
f'{res_num}_') # "_" selects residue with no insertion codes
else:
clean_res_num.append(str(res_num))
clean_protein = p.select('resnum {}'.format(' '.join(clean_res_num)))
pdb_name = os.path.basename(pdb_file).split('.')[0]
if write_out:
pdy.writePDB(f'./pdb_clean_new/{pdb_name}_{chain}.pdb', clean_protein)
# Select cleaned residues
return clean_protein
def print_cluster(self, mems, label, outdir, number=None):
"""prints pdbs of the top number of hotspots (number) to the output directory (outdir)."""
if outdir[-1] != '/':
outdir += '/'
if number:
mems = random.sample(mems, number)
for mem in mems:
resnum_chid = tuple(self._all_resnum_chid[mem])
type_ = self._all_type[mem]
typestr = type_
resn = self._all_resn[mem]
vdm_tags = self._all_vdm_tags[mem]
if typestr == 'PHI_PSI':
typestr = 'PHI_PSI/' + self.rel_vdm_phipsi_bin[resnum_chid]
pdbpath = self.rel_vdm_path + typestr + '/pdbs/' + resn + '/'
filename = 'iFG_' + str(vdm_tags[0]) + '_vdM_' + str(vdm_tags[1]) + '_iFlip_' \
+ str(vdm_tags[2]) + '_' + self.name + '_' + 'oriented.pdb.gz'
pdb = pr.parsePDB(pdbpath + filename)
old_coords = pdb.getCoords()
new_coords = \
np.dot((old_coords - self._rois_rot_trans[resnum_chid][type_][resn][1]),
self._rois_rot_trans[resnum_chid][type_][resn][0]) \
+ self._rois_rot_trans[resnum_chid][type_][resn][2]
pdb.setCoords(new_coords)
newfile_path = outdir + 'cluster/' + label + '/'
newfile_name = self.name + '_cluster_' + label + '_' \
+ ''.join(str(x) for x in resnum_chid) + '_' + type_ + '_' + filename
try:
os.makedirs(newfile_path)
except:
pass
pr.writePDB(newfile_path + newfile_name, pdb)
def print_hotspots(self, outdir, number=25):
"prints pdbs of the top number of hotspots (number) to the output directory (outdir)."
if outdir[-1] != '/':
outdir += '/'
for i in range(number):
for resnum_chid, index in self.hotspots[i]:
resn = self._resn[resnum_chid][index]
type_ = self._type[resnum_chid][index]
typestr = type_
vdm_tags = self._vdm_tags[resnum_chid][index]
if typestr == 'PHI_PSI':
typestr = 'PHI_PSI/' + self.rel_vdm_phipsi_bin[resnum_chid]
pdbpath = self.rel_vdm_path + typestr + '/pdbs/' + resn + '/'
filename = 'iFG_' + str(vdm_tags[0]) + '_vdM_' + str(vdm_tags[1]) + '_iFlip_' \
+ str(vdm_tags[2]) + '_' + self.name + '_' + 'oriented.pdb.gz'
pdb = pr.parsePDB(pdbpath + filename)
old_coords = pdb.getCoords()
new_coords = \
np.dot((old_coords - self._rois_rot_trans[resnum_chid][type_][resn][1]),
self._rois_rot_trans[resnum_chid][type_][resn][0]) \
+ self._rois_rot_trans[resnum_chid][type_][resn][2]
pdb.setCoords(new_coords)
newfile_path = outdir + 'hotspots/' + str(i + 1) + '/'
newfile_name = self.name + '_hotspot_' + str(i + 1) + '_' \
+ ''.join(str(x) for x in resnum_chid) + '_' + type_ + '_' + filename
try:
os.makedirs(newfile_path)
except:
pass
pr.writePDB(newfile_path + newfile_name, pdb)
def preprocess_single(model, chain_name):
hv = model.getHierView()
for chain in hv.iterChains():
chain.setChids(chain_name)
prody.writePDB("tmp.pdb", model)
model = prody.parsePDB("tmp.pdb")
check_call(["rm", "tmp.pdb"])
pos = 1
apos = 1
hv = model.getHierView()
for chain in hv.iterChains():
# print "chain : " , chain
for res in chain.iterResidues():
res.setResnums(pos)
for atom in res:
atom.setSerial(apos)
apos += 1
if len(res.getIcode()) == 0:
pos += 1
hv.update()
return model
def print_hotspot(self, hs_subgr_num, outdir):
"""prints pdbs of the top number of hotspots (number) to the output directory (outdir)."""
if outdir[-1] != '/':
outdir += '/'
for node in self.hotspot_subgraphs[hs_subgr_num]:
resnum_chid = node[0]
type_ = self.hotspot_subgraphs[hs_subgr_num].node[node]['type']
typestr = type_
resn = self.hotspot_subgraphs[hs_subgr_num].node[node]['resn']
vdm_tags = self.hotspot_subgraphs[hs_subgr_num].node[node][
'vdm_tags']
if typestr == 'PHI_PSI':
typestr = 'PHI_PSI/' + self.rel_vdm_phipsi_bin[resnum_chid]
pdbpath = self.rel_vdm_path + typestr + '/pdbs/' + resn + '/'
filename = 'iFG_' + str(vdm_tags[0]) + '_vdM_' + str(vdm_tags[1]) + '_iFlip_' \
+ str(vdm_tags[2]) + '_' + self.name + '_' + 'oriented.pdb.gz'
pdb = pr.parsePDB(pdbpath + filename)
old_coords = pdb.getCoords()
new_coords = \
np.dot((old_coords - self._rois_rot_trans[resnum_chid][type_][resn][1]),
self._rois_rot_trans[resnum_chid][type_][resn][0]) \
+ self._rois_rot_trans[resnum_chid][type_][resn][2]
pdb.setCoords(new_coords)
newfile_path = outdir + 'hotspots/' + str(hs_subgr_num + 1) + '/'
newfile_name = self.name + '_hotspot_' + str(hs_subgr_num + 1) + '_' \
+ ''.join(str(x) for x in resnum_chid) + '_' + type_ + '_' + filename
try:
os.makedirs(newfile_path)
except:
pass
pr.writePDB(newfile_path + newfile_name, pdb)
def PDB_creator(self, peptides: list, overlaps: list):
res = er.PepExtractor(database=self.db, resfile=self.resfile)
last_pep = peptides[0][:-1]
last_conform = peptides[0][-1]
r1 = res.extract_result(str(last_pep), int(last_conform))
seq = last_pep
for i in range(1, len(peptides)):
last_pep = peptides[i][:-1]
last_conform = peptides[i][-1]
last_overlap = overlaps[i - 1]
r2 = res.extract_result(str(last_pep), int(last_conform))
r1_ = r1.select(
'resnum 2:{0}'.format(len(seq) + 2)).toAtomGroup()
r2_ = r2.select('resnum ' + str(last_overlap + 2) +
':5').toAtomGroup()
c_ = 5
for r_ in r2_.iterResidues():
r_.setResnum(c_)
c_ += 1
r1 = r1_ + r2_
seq += last_pep[last_overlap:]
prody.writePDB(
'{0}{1}{2}.pdb'.format(self.outdir, str(len(seq)), seq),
r1)
def main():
in_file, trajectory, out_file, force_const = parse_args()
average = load_pdb(in_file)
trajectory = load_pdb(trajectory)
starting_model = get_closest_frame(trajectory, average)
minimized_protein = run_minimization(average, starting_model, force_const)
prody.writePDB(out_file, minimized_protein)
def save_representatives(
representatives,
pdb_name,
workspace_handler,
trajectory_holder,
do_merged_files_have_correlative_models,
write_frame_number_instead_of_correlative_model_number,
keep_remarks=False):
"""
Saves a pdb file containing the most representative elements of the clustering.
@param representatives: A list of the representative elements of the clustering we want to extract.
@param workspace_handler: The workspace handler of this run.
@param trajectory_holder: The trajectory handler for this run or an array with pdb file paths.
@param do_merged_files_have_correlative_models: When merging, output file will have models from 0 to M, where M is the total number
of frames of the merged file.
@param write_frame_number_instead_of_model_number: When extracting frames, extract those models which number coincides with the
frame numbers in 'representatives'. Otherwise, extract those models which position coincide with the frame number in
'representatives'.
"""
results_directory = workspace_handler["results"]
# Merge pdbs (in order)
temporary_merged_trajectory_path = os.path.join(
workspace_handler["tmp"], "tmp_merged_trajectory.pdb")
#===========================================================
# THIS DOES NOT WORK IF USING DCD FILES
# merge_pdbs(trajectory_holder,
# temporary_merged_trajectory_path,
# do_merged_files_have_correlative_models)
# TEMPORARY HACK TO OVERCOME DCD MERGING BUG
merged_pdb = trajectory_holder.getMergedStructure()
prody.writePDB(temporary_merged_trajectory_path, merged_pdb)
#==========================================================
# Extract frames from the merged pdb
file_handler_in = open(temporary_merged_trajectory_path, "r")
file_handler_out = open(
os.path.join(results_directory, "%s.pdb" % pdb_name), "w")
pdb_tools.extract_frames_from_trajectory_sequentially(
file_handler_in=file_handler_in,
number_of_frames=pdb_tools.get_number_of_frames(
temporary_merged_trajectory_path),
file_handler_out=file_handler_out,
frames_to_save=representatives,
write_frame_number_instead_of_correlative_model_number=
write_frame_number_instead_of_correlative_model_number,
keep_header=keep_remarks)
file_handler_in.close()
file_handler_out.close()
return os.path.join(results_directory, "%s.pdb" % pdb_name)
def _generate_sidechains_scwrl(self):
if not self.rec is None:
rec = self.rec.copy()
rec.setChids('A')
lig = self._tpl.copy()
rec = BasePDB(ag=rec)
lig = BasePDB(ag=lig)
merged = rec.add_mol(lig, keep_resi=False, keep_chains=True)
merged.save(self._mrg_file)
self._make_scwrl_sequence_file()
call = [
define.SCWRL_EXE, '-h', '-i', self._mrg_file, '-o',
self._scw_file, '-s', self._seq_file
]
else:
prody.writePDB(self._mrg_file, self._tpl)
call = [
define.SCWRL_EXE, '-h', '-i', self._mrg_file, '-o',
self._scw_file
]
# scwrl wants rosetta hydrogen naming
BasePDB(self._mrg_file).to_rosetta().save(self._mrg_file)
helpers.shell_call(call)
pep = prody.parsePDB(self._scw_file)
# extract peptide and renumber
pep = BasePDB(ag=pep.select('chain B').copy()).renumber(
keep_resi=False).ag
self.pep = pep
def preprocess_single( model, chain_name ): hv = model.getHierView() for chain in hv.iterChains(): chain.setChids( chain_name ) prody.writePDB( 'tmp.pdb', model ) model = prody.parsePDB( 'tmp.pdb' ) check_call( ["rm", "tmp.pdb"] ) pos=1 apos=1 hv = model.getHierView() for chain in hv.iterChains(): #print "chain : " , chain for res in chain.iterResidues(): res.setResnums( pos ) for atom in res: atom.setSerial( apos ) apos += 1 if len( res.getIcode() ) == 0: pos += 1 hv.update() return model
def savePDB(pdb, filename):
"""
Saves a prody pdb data structure in a pdb format file (simple wrapper).
@param pdb: A prody pdb data structure.
"""
prody.writePDB(filename, pdb)
def phi2pdb(base_pdb, phi, save_path="./"):
pdb = prody.parsePDB(base_pdb)
atoms = [a for a in pdb]
for a in pdb:
a.setBeta(0.0)
phif = open(phi, "r")
phic = phif.readlines()
phif.close()
for l in phic:
x,y,z,k = l.strip().split(",")
x = np.float(x)
y = np.float(y)
z = np.float(z)
k = np.float(k)
for a in xrange(len(atoms)):
X,Y,Z = atoms[a].getCoords()
if X==x and Y==y and Z==z:
atoms[a].setBeta(k)
atoms.pop(a)
break
out_pdb = os.path.join(save_path, phi.split(os.sep)[-1][:-4]+".pdb")
prody.writePDB(out_pdb, pdb)
return out_pdb
def main():
args = parse_args()
trajectory = load_pdb(args.trajectory)
average = calc_average(trajectory)
prody.writePDB(args.out_file, average)
def _load_traj(ag):
h, tmp = tempfile.mkstemp(dir='.', suffix='.pdb')
os.close(h)
prody.writePDB(tmp, ag)
traj = md.load_pdb(tmp, frame=0)
os.remove(tmp)
return traj
def projectCentralitiesOntoProteinPyMol(centrality, centralityArray, out_file, \
selectedAtoms, scalingFactor):
"""
Produces PyMol output files for visualizing protein centralities.
This function writes a pml file and a PDB file that can be viewed in
VMD. Bfactor field of the protein contains the centrality information.
The first N residues with the highest centrality are highlighed in VDW
representation. that that contains the centralities on
on Bfactor field of the pdb.
The output files can be visualized with VMD (Visual Molecular
dynamics) program as follows: pymol output.pml
Parameters
----------
centrality: string
It can have 'degree', 'betweenness', 'closeness',
'current_flow_betweenness' or 'current_flow_closeness'.
centralityArray: A numpy data array ?
It is a numpy matrix of typically nDCC, LMI or Generalized Correlations.
out_file: string
Prefix of the output file. According to the centralty measure, it will be
extended.
selectedAtoms: object
This is a prody.parsePDB object of typically CA atoms of a protein.
ScalingFactor: float
Sometimes, the values of the centrality arrays are too small.
The scaling factor multiplies the array to make the values visible in
the Bfactor colums.
Returns
-------
Nothing
"""
percentage = 0.10
numKeyResidues = int(percentage * len(selectedAtoms))
PML_FILE = open(out_file + '_' + centrality + '.pml', 'w')
PML_FILE.write("load " + out_file + "_" + centrality + ".pdb" + "\n")
PML_FILE.write("cartoon type = tube\n")
PML_FILE.write("spectrum b\n")
PML_FILE.write("set sphere_scale, 0.75\n\n")
vdw_representation_string = "show spheres, chain {0:s} and resi {1:d} and name ca\n"
sortedList = np.flip(np.argsort(centralityArray))
for i in range(0, numKeyResidues):
# print(centralityArray[sortedList[i]])
PML_FILE.write(vdw_representation_string.\
format(selectedAtoms.getChids()[sortedList[i]],
selectedAtoms.getResnums()[sortedList[i]]))
selectedAtoms.setBetas([scalingFactor * i for i in centralityArray])
writePDB(out_file + '_' + centrality + '.pdb', selectedAtoms)
PML_FILE.close()
def main():
in_file, trajectory, out_file, skip_first = parse_args()
average = load_pdb(in_file)
trajectory = load_pdb(trajectory)
starting_model = get_closest_frame(trajectory,
average,
skip_first=skip_first)
prody.writePDB(out_file, starting_model)
def make_interactamers(df, an, outpath, comb):
if outpath[-1] != '/':
outpath += '/'
resns = set(df.groupby('resname_vdm').groups).intersection(
set(interactamer_atoms.keys()))
for resn in resns:
for key in interactamer_atoms[resn].keys():
origin_atom, plane_atom1, plane_atom2 = interactamer_atoms[resn][
key]
vdms = parse_interactamers_aa(df, an, resn)
if vdms:
pdb_path = outpath + 'pdbs/' + resn + '/' + key + '/'
try:
os.makedirs(pdb_path)
except:
pass
picklepath = outpath + 'pickle/'
try:
os.makedirs(picklepath)
except:
pass
rel_vdm_output = []
print('Making interactamer vdMs for ' + resn + ', ' + key +
'...')
for vdm in vdms:
try:
ifg_count = int(repr(vdm).split('_')[1])
vdm_count = int(repr(vdm).split('_')[3])
# if resn == 'PRO':
# bb_coords, sc_coords, ifg_coords, pdbs = make_rel_vdm_coords(vdm, comb, origin_atom,
# 'CD', plane_atom2)
# else:
bb_coords, sc_coords, sc_coords_ON, sc_coords_CS, ifg_coords, ifg_ON_coords, ifg_CS_coords, pdbs = make_rel_vdm_coords(
vdm, comb, origin_atom, plane_atom1, plane_atom2)
for i, bbc, scc, sccon, scccs, ic, icon, iccs, pdb in zip(
range(len(bb_coords)), bb_coords, sc_coords,
sc_coords_ON, sc_coords_CS, ifg_coords,
ifg_ON_coords, ifg_CS_coords, pdbs):
rel_vdm_output.append([
ifg_count, vdm_count, i + 1, bbc, scc, sccon,
scccs, ic, icon, iccs, resn
])
string = repr(pdb).split()[1].split('_')
pr.writePDB(
pdb_path + '_'.join(string[:-1]) + '_iFlip_' +
str(i + 1) + '_' + string[-1] +
'_oriented.pdb.gz', pdb)
except Exception:
traceback.print_exc()
# output format = [ifg_count, vdm_count, ifg_flip, bb_coords, sc_coords_all, sc_coords_ON, sc_coords_CS, ifg_coords, ifg_ON_coords, ifg_CS_coords, vdm_resn]
if rel_vdm_output:
with open(picklepath + resn + '_rel_vdms.pickle',
'wb') as f:
pickle.dump(np.array(rel_vdm_output, dtype=object), f)
else:
shutil.rmtree(pdb_path)
def write_PDB(self, structure, pdb_name):
"""
Write a prody protein to a pdb file
:param protein: protein object from prody
:param pdb_name: base name for the pdb file
:return: None
"""
prody.writePDB(f"{pdb_name}", structure)
def write_superposed_pdbs(self, output_pdb_folder, alignments: dict = None):
"""
Superposes PDBs according to alignment and writes transformed PDBs to files
(View with Pymol)
Parameters
----------
alignments
output_pdb_folder
"""
if alignments is None:
alignments = self.alignment
output_pdb_folder = Path(output_pdb_folder)
if not output_pdb_folder.exists():
output_pdb_folder.mkdir()
reference_name = self.structures[0].name
reference_pdb = pd.parsePDB(
str(self.output_folder / f"cleaned_pdb/{self.structures[0].name}.pdb")
)
core_indices = np.array(
[
i
for i in range(len(alignments[reference_name]))
if -1 not in [alignments[n][i] for n in alignments]
]
)
aln_ref = alignments[reference_name]
ref_coords_core = (
reference_pdb[helper.get_alpha_indices(reference_pdb)]
.getCoords()
.astype(np.float64)[np.array([aln_ref[c] for c in core_indices])]
)
ref_centroid = helper.nb_mean_axis_0(ref_coords_core)
ref_coords_core -= ref_centroid
transformation = pd.Transformation(np.eye(3), -ref_centroid)
reference_pdb = pd.applyTransformation(transformation, reference_pdb)
pd.writePDB(str(output_pdb_folder / f"{reference_name}.pdb"), reference_pdb)
for i in range(1, len(self.structures)):
name = self.structures[i].name
pdb = pd.parsePDB(
str(self.output_folder / f"cleaned_pdb/{self.structures[i].name}.pdb")
)
aln_name = alignments[name]
common_coords_2 = (
pdb[helper.get_alpha_indices(pdb)]
.getCoords()
.astype(np.float64)[np.array([aln_name[c] for c in core_indices])]
)
(
rotation_matrix,
translation_matrix,
) = superposition_functions.svd_superimpose(
ref_coords_core, common_coords_2
)
transformation = pd.Transformation(rotation_matrix.T, translation_matrix)
pdb = pd.applyTransformation(transformation, pdb)
pd.writePDB(str(output_pdb_folder / f"{name}.pdb"), pdb)
def main():
r = parse_args()
input_pdb_filename = r[0]
trajectory_filename = r[1]
output_pdb_filename = r[2]
force_const = r[3]
m = Minimizer(input_pdb_filename, trajectory_filename)
minimized_protein = m.run_minimization(force_const)
prody.writePDB(output_pdb_filename, minimized_protein)
def ligand_parse_write(path, out, lig_name):
pose = parsePDB(path)
writePDB(out + 'lig_' + path.split('/')[-1],
pose.select('hetero and resname {}'.format(lig_name)))
ligand_path = out + 'lig_' + path.split('/')[-1]
return ligand_path
def save(self, pdb, **kwargs):
if not pdb.endswith('.pdb'):
raise RuntimeError('Filename must end with .pdb')
ag = self.ag.copy()
new_names = ['%4s' % ('%-3s' % a) if not ('0' <= a[0] <= '9') else '%-4s' % a for a in ag.getNames()]
ag.setNames(new_names)
prody.writePDB(pdb, ag, **kwargs)
return pdb
def main():
from argparse import ArgumentParser
parser = ArgumentParser()
parser.add_argument("--ftfile-line",
"-n",
type=int,
default=0,
help="0 indexed")
parser.add_argument("--limit",
"-l",
type=int,
default=None,
help="limit for number of ft lines read")
parser.add_argument("ligand", help="ligand used for docking")
parser.add_argument("ft_file",
help="ftresults file to draw transformations from.")
parser.add_argument(
"rot_file",
help="Rotation file used during PIPER run to make ft_file.")
parser.add_argument("--ft_limit",
help="Speficy how many ftfile lines to read")
args = parser.parse_args()
rotations = read_rotations(args.rot_file)
#read ft_files
ftresults = read_ftresults(args.ft_file, args.limit)
#get center information for original ligands
lig_orig = parsePDB(args.ligand)
lig_center = np.mean(lig_orig.getCoords(), axis=0)
lig_name = args.ligand
lig_base = lig_name.rsplit('/', 1)[-1][:-4]
#apply ft lines to the aligned ligands
coords_apply = apply_ftresults_atom_group(lig_orig,
ftresults,
rotations,
center=lig_center)
#create new atom group for translated ligand
lig_new = AtomGroup('%s.%s.pdb' % (lig_base, args.ftfile_line))
#only choose one coordinate set (0 indexed)
coords_apply.setACSIndex(args.ftfile_line)
lig_new.setCoords(coords_apply.getCoords())
lig_new.setNames(lig_orig.getNames())
lig_new.setResnames(lig_orig.getResnames())
lig_new.setResnums(lig_orig.getResnums())
#write new ligand file
writePDB("%s.%s.pdb" % (lig_base, args.ftfile_line), lig_new)
def split_receptor(bucket, table_idx, param,
datum): # todo (maksym) param = params;
try: # todo (maksym) datum = pdb_name
if type(datum).__name__ in ['tuple', 'list']:
datum = datum[0]
receptor = datum # todo receptor = pdb_name !!!!!
output_folder = param['output_folder']
output_folder = '{}_{}'.format(table_idx, output_folder)
input_download_folder = param['input_download_folder']
input_pdb_dir = os.path.join(data_dir, input_download_folder)
input_pdb_path = os.path.join(input_pdb_dir, receptor + '.pdb')
parsed_pdb = prody.parsePDB(input_pdb_path)
parsed_header = prody.parsePDBHeader(input_pdb_path)
output_rec_dir = os.path.join(data_dir, output_folder, receptor)
_makedir(output_rec_dir)
ligands = []
for chem in parsed_header['chemicals']:
chain, resnum, resname = chem.chain, chem.resnum, chem.resname
ligands.append([chain, str(resnum), resname])
for chain, resnum, resname in ligands:
try:
rec = parsed_pdb.select('not (chain {} resnum {})'.format(
chain, resnum))
rec = rec.select('not water')
heavy_atom = rec.select('not hydrogen').numAtoms()
rec_name = '_'.join(
[receptor, chain, resnum, resname, 'receptor']) + '.pdb'
prody.writePDB(os.path.join(output_rec_dir, rec_name), rec)
record = [
receptor, chain, resnum, resname, heavy_atom,
parsed_header['experiment'], parsed_header['resolution'],
1, 'success'
]
records = [record]
db.insert(table_idx, records, bucket=bucket)
except Exception as e:
record = [
receptor, chain, resnum, resname, 0, 0, 0, 0,
str(e)
] # datum = failure_message
records = [record]
print(records)
db.insert(table_idx, records, bucket=bucket)
# TODO: (maksym) I believe this is controllable with logging
except Exception as e:
print(e)
raise Exception(str(e))
def get_pocket(input_path):
"""
select protein atoms from pocket file
"""
output_path = input_path.replace(config.input_data_dir, config.data_dir)
if not os.path.exists(output_path):
parsed = prody.parsePDB(pocket_input)
pocket = parsed.select('protein')
_mkdir(os.path.dirname(output_path))
prody.writePDB(output_path, pocket)
def write_pdb(structure, file_name='file.pdb'):
"""Takes a Prody structure prody.AtomGroup and writes it on a pdb file
Parameters
-------------
strucure : prody.AtomGroup
file_name : str
default "file.pdb"
"""
prody.writePDB(file_name, structure)
def writepdb_and_get_pickle_info(pdb_path, pklout, vdm, comb, origin_atom,
plane_atom1, plane_atom2, resn, ifg_count,
vdm_count):
'''helper function for make_rel_vdms and make_interactamers.
Need vdmelem1dists to choose which vdm flipping to keep.'''
if resn != 'TYR' and resn not in flip_residues:
bb_coords, sc_coords, sc_coords_ON, sc_coords_CS, ifg_coords, ifg_ON_coords, ifg_CS_coords, pdbs, vdmelem1avgdists = \
make_rel_vdm_coords(vdm, comb, origin_atom, plane_atom1, plane_atom2)
elif resn == 'TYR':
x = make_rel_vdm_coords(vdm, comb, origin_atom, plane_atom1,
plane_atom2)
y = make_rel_vdm_coords(vdm,
comb,
origin_atom,
'CE2',
plane_atom2,
unflipped=False)
xelem1dist = x[8]
yelem1dist = y[8]
if xelem1dist <= yelem1dist:
bb_coords, sc_coords, sc_coords_ON, sc_coords_CS, ifg_coords, ifg_ON_coords, ifg_CS_coords, pdbs, vdmelem1avgdists = x
else:
bb_coords, sc_coords, sc_coords_ON, sc_coords_CS, ifg_coords, ifg_ON_coords, ifg_CS_coords, pdbs, vdmelem1avgdists = y
elif resn in flip_residues:
x = make_rel_vdm_coords(vdm, comb, origin_atom, plane_atom1,
plane_atom2)
y = make_rel_vdm_coords(vdm,
comb,
origin_atom,
plane_atom2,
plane_atom1,
unflipped=False)
xelem1dist = x[8]
yelem1dist = y[8]
if xelem1dist <= yelem1dist:
bb_coords, sc_coords, sc_coords_ON, sc_coords_CS, ifg_coords, ifg_ON_coords, ifg_CS_coords, pdbs, vdmelem1avgdists = x
else:
bb_coords, sc_coords, sc_coords_ON, sc_coords_CS, ifg_coords, ifg_ON_coords, ifg_CS_coords, pdbs, vdmelem1avgdists = y
for i, bbc, scc, sccon, scccs, ic, icon, iccs, pdb in zip(range(len(bb_coords)), bb_coords, sc_coords, \
sc_coords_ON, sc_coords_CS, ifg_coords, ifg_ON_coords, ifg_CS_coords, pdbs):
pklout.append([
ifg_count, vdm_count, bbc, scc, sccon, scccs, ic, icon, iccs, resn
])
string = repr(pdb).split()[1].split('_')
pr.writePDB(
pdb_path + '_'.join(string[:-1]) + '_' + string[-1] +
'_oriented.pdb.gz', pdb)
return pklout
def write_pdb(structure, file_name="file.pdb"):
"""
Takes a Prody strucure prody.AtomGroup and writes it on a pdb file
strucure :: prody.AtomGroup
file_name :: string, default "file.pdb"
returns nothing
"""
prody.writePDB(file_name, structure)
def save_all_clusters(my_params, pdb_params, workspaceHandler, trajectoryHandler, clustering, generatedFiles, timer):
timer.start("Save clusters")
#Parameters
keep_remarks = my_params["keep_remarks"] if "keep_remarks" in my_params else False
keep_frame_number = my_params["keep_frame_number"] if "keep_frame_number" in my_params else False
# Places
results_place = workspaceHandler["results"]
clusters_place = workspaceHandler["clusters"]
tmp_place = workspaceHandler["tmp"]
#===========================================================
# THIS DOES NOT WORK IF USING DCD OR STRUCTS
# The real job
# input_path = os.path.join(tmp_place, "tmp_merged_trajectory.pdb")
# merge_pdbs(pdb_params, input_path)
# TEMPORARY HACK TO OVERCOME DCD MERGING BUG
merged_pdb = trajectoryHandler.getMergedStructure()
input_path = os.path.join(tmp_place, "tmp_merged_trajectory.pdb")
prody.writePDB(input_path, merged_pdb)
#==========================================================
number_of_frames = get_number_of_frames(input_path)
cluster_files = []
for cluster in clustering.clusters:
output_path = os.path.join(clusters_place, "%s.pdb"%(cluster.id))
cluster_files.append(output_path)
file_handler_in = open(input_path,"r")
file_handler_out = open(output_path,"w")
extract_frames_from_trajectory_sequentially(file_handler_in,
number_of_frames,
file_handler_out,
cluster.all_elements,
keep_header = keep_remarks,
write_frame_number_instead_of_correlative_model_number=keep_frame_number)
file_handler_in.close()
file_handler_out.close()
# Add all bz2 files to a tar file
tar_path = os.path.join(results_place,"clusters.tar.gz")
tar = tarfile.open(tar_path, "w:gz")
for comp_file in cluster_files:
tar.add(comp_file, os.path.basename(comp_file))
tar.close()
timer.stop("Save clusters")
generatedFiles.append({"description":"Clusters",
"path":os.path.abspath(tar_path),
"type":"compressed_pdb"})
def _find_ligand(self):
self._get_file_path(ligand=True)
protein = parsePDB(self.file_path)
try:
seq = protein['A'].getSequence()
except:
pass
else:
ligand = protein.select('not protein and not water')
repr(ligand)
if ligand:
self.out_filename = self.file_path.split('.')[0] + '_ligand.pdb'
writePDB(self.out_filename, ligand)
def get_protein(input_path):
"""
select protein atoms from protein file
"""
try:
output_path = input_path.replace(config.input_data_dir, config.data_dir)
if not os.path.exists(output_path):
parsed = prody.parsePDB(input_path)
pocket = parsed.select('protein')
_mkdir(os.path.dirname(output_path))
prody.writePDB(output_path, pocket)
except Exception as e:
print e
def split(receptor, pdb_outpath, init='split_init'):
init = eval(init)
rec_dir = os.path.join(init.data_dir, init.rec_folder)
lig_dir = os.path.join(init.data_dir, init.lig_folder)
pdb_path = os.path.join(init.data_dir, pdb_outpath)
parsed_pdb = prody.parsePDB(pdb_path)
parsed_header = prody.parsePDBHeader(pdb_path)
ligands = []
for chem in parsed_header['chemicals']:
ligands.append([chem.chain, str(chem.resnum), chem.resname])
splited = []
for chain, resnum, resname in ligands:
lig = parsed_pdb.select('chain {} resnum {}'.format(chain, resnum))
rec = parsed_pdb.select('not (chain {} resnum {})'.format(
chain, resnum))
if lig is None:
continue
resid = lig.getHierView().iterResidues().next().getResindex()
resid = str(resid)
heavy_lig = lig.select('not hydrogen')
heavy_atom = heavy_lig.numAtoms()
heavy_coord = heavy_lig.getCoords()
#max_size_on_axis = max(heavy_coord.max(axis=0) - heavy_coord.min(axis=0))
#Changing max_size_on_axis to max pairwise distance between coords
#max_size_on_axis = max(scipy.spatial.distance.pdist(heavy_coord).tolist())
lig_name = '_'.join([receptor, chain, resnum, resname, 'ligand'
]) + '.pdb'
if not os.path.exists(os.path.join(lig_dir, receptor)):
os.makedirs(os.path.join(lig_dir, receptor))
prody.writePDB(os.path.join(lig_dir, receptor, lig_name), lig)
rec_name = '_'.join([receptor, chain, resnum, resname, 'receptor'
]) + '.pdb'
if not os.path.exists(os.path.join(rec_dir, receptor)):
os.makedirs(os.path.join(rec_dir, receptor))
prody.writePDB(os.path.join(rec_dir, receptor, rec_name), rec)
splited.append([
receptor,
str(resname),
os.path.join(init.rec_folder, receptor, rec_name),
os.path.join(init.lig_folder, receptor, lig_name)
])
return splited
def write_pdb_dir(pdb_dir, struct_data):
# import prody for pdb read/write
import prody
# create output directory, if not yet present
if not(os.path.exists(pdb_dir)):
os.makedirs(pdb_dir)
# write protein chain pdb files to output directory
for (pdb_f, struct) in struct_data:
if not(struct is None):
out_f = os.path.join(pdb_dir, pdb_f)
prody.writePDB(out_f, struct)
def prody_contacts(**kwargs):
"""Identify contacts of a target structure with one or more ligands.
Contacting atoms (or extended subset of atoms, such as residues) are
outputted in PDB file format.
:arg target: target PDB identifier or filename
:arg ligand: ligand PDB identifier(s) or filename(s)
:arg select: atom selection string for target structure
:arg radius: contact radius (Å), default is ``4.0``
:arg extend: output same ``'residue'``, ``'chain'``, or ``'segment'`` along
with contacting atoms
:arg prefix: prefix for output file, default is *target* filename
:arg suffix: output filename suffix, default is *ligand* filename"""
import prody
LOGGER = prody.LOGGER
target = prody.parsePDB(kwargs['target'])
title = kwargs.get('prefix') or target.getTitle()
selstr = kwargs.get('select')
if selstr:
target = target.select(selstr)
contacts = prody.Contacts(target)
suffix = kwargs.get('suffix', '_contacts')
extend = kwargs.get('extend')
radius = float(kwargs.get('radius', 4.0))
ligands = kwargs.get('ligand')
if len(ligands) > 1:
outfn = lambda fn: title + suffix + '_' + fn + '.pdb'
else:
outfn = lambda fn: title + suffix + '.pdb'
for pdb in ligands:
ligand = prody.parsePDB(pdb)
sel = contacts(radius, ligand)
if sel:
LOGGER.info('{0} atoms from {1} contact {2}.'
.format(len(sel), pdb, str(target)))
if extend:
sel = target.select('same ' + extend + ' as sel', sel=sel)
LOGGER.info('Selection is extended to {0} atoms of the same '
'{1}(s).'.format(len(sel), extend))
pdbfn = outfn(ligand.getTitle())
LOGGER.info('Writing contacts into ' + pdbfn)
prody.writePDB(pdbfn, sel)
def prody_select(opt):
"""Write selected atoms from a PDB file in PDB format."""
import prody
LOGGER = prody.LOGGER
pdb = prody.parsePDB(opt.pdb)
prefix = opt.output
if not prefix:
prefix = pdb.getTitle() + '_selected'
pdbselect = pdb.select(opt.selstr)
if pdbselect is None:
opt.subparser('Selection "{0:s}" do not match any atoms.'
.format(opt.selstr))
LOGGER.info('Writing ' + prefix + '.pdb')
prody.writePDB(prefix + '.pdb', pdbselect)
def prody_align(opt):
"""Align models in a PDB file or a PDB file onto others."""
import prody
LOGGER = prody.LOGGER
args = opt.pdb
if len(args) == 1:
pdb = args[0]
LOGGER.info('Aligning multiple models in: ' + pdb)
selstr, prefix, model = opt.select, opt.prefix, opt.model
pdb = prody.parsePDB(pdb)
pdbselect = pdb.select(selstr)
if pdbselect is None:
opt.subparser.error('Selection {0:s} do not match any atoms.'
.format(repr(selstr)))
LOGGER.info('{0:d} atoms will be used for alignment.'
.format(len(pdbselect)))
pdbselect.setACSIndex(model-1)
prody.printRMSD(pdbselect, msg='Before alignment ')
prody.alignCoordsets(pdbselect)
prody.printRMSD(pdbselect, msg='After alignment ')
if prefix == '':
prefix = pdb.getTitle() + '_aligned'
outfn = prefix + '.pdb'
LOGGER.info('Writing file: ' + outfn)
prody.writePDB(outfn, pdb)
else:
reffn = args.pop(0)
seqid=opt.seqid
overlap=opt.overlap
LOGGER.info('Aligning structures onto: ' + reffn)
ref = prody.parsePDB(reffn)
for arg in args:
if arg == reffn:
continue
if '_aligned.pdb' in arg:
continue
pdb = prody.parsePDB(arg)
result = prody.matchAlign(pdb, ref, seqid=seqid, overlap=overlap,
tarsel=opt.select, allcsets=True,
cslabel='Model', csincr=1)
if result:
outfn = pdb.getTitle() + '_aligned.pdb'
LOGGER.info('Writing file: ' + outfn)
prody.writePDB(outfn, pdb)
else:
LOGGER.warning('Failed to align ' + arg)
def fix_openmm():
# get the whole crystal structure
# get only the ATOM records
# and HETAM records for MSE
# convert MSE to MET
with open('no_smet.pdb', 'w') as outfile:
with open('experimental.pdb') as infile:
for line in infile:
if line.startswith('ATOM'):
outfile.write(line)
if line.startswith('HETATM'):
if line[17:20] == 'MSE':
atom_name = line[12:17]
if atom_name == 'SE ':
atom_name = ' SD '
line_fixed = 'ATOM ' + line[6:12] + atom_name + 'MET' + line[20:67] + '\n'
outfile.write(line_fixed)
# load the file into prody
p = prody.parsePDB('no_smet.pdb')
p = p.select('not hydrogen')
# get one of the rosetta models
r = prody.parsePDB('rosetta.pdb')
# perform an alignment to find out what part of the crystal structure
# corresponds to the rosetta file
match = prody.matchChains(r, p, subset='all', overlap=25, pwalign=True)[0][1]
print len(match)
prody.writePDB('chain.pdb', match)
# now clean it up with pdb fixer
subprocess.check_call('python ~/Source/PdbFixer/pdbfixer.py chain.pdb', shell=True)
# now load it with zam
p = protein.Protein('output.pdb')
p.Dehydrogen()
disulfide_pairs = find_disulfide(p)
for r1, r2 in disulfide_pairs:
print ' added disulfide between {} and {}'.format(r1, r2)
p.Res[r1].FullName = 'CYX'
p.Res[r2].FullName = 'CYX'
p.WritePdb('start.pdb')
# now run tleap
print ' running tleap'
run_tleap(disulfide_pairs)
def combine_structures(directory_with_pdbs, output_filename):
search_string = join(directory_with_pdbs, '*.pdb')
path_list = glob(search_string)
if len(path_list) > MAX_FRAMES:
raise RuntimeError('Got %d frames, but only up to %d frames are allowed.')
else:
pass
atom_group = parsePDB(path_list[0])
for i, path in enumerate(path_list):
if i == 0:
continue
else:
p = parsePDB(path)
atom_group.addCoordset(p)
writePDB(output_filename, atom_group)
def prody_align(opt):
"""Align models in a PDB file or a PDB file onto others."""
import prody
LOGGER = prody.LOGGER
args = opt.pdb
if len(args) == 1:
pdb = args[0]
LOGGER.info('Aligning multiple models in: ' + pdb)
selstr, prefix, model = opt.select, opt.prefix, opt.model
pdb = prody.parsePDB(pdb)
pdbselect = pdb.select(selstr)
if pdbselect is None:
LOGGER.warning('Selection "{0:s}" do not match any atoms.'
.format(selstr))
sys.exit(-1)
LOGGER.info('{0:d} atoms will be used for alignment.'
.format(len(pdbselect)))
pdb.setACSIndex(model-1)
prody.alignCoordsets(pdb, selstr=selstr)
rmsd = prody.calcRMSD(pdb)
LOGGER.info('Max RMSD: {0:0.2f} Mean RMSD: {1:0.2f}'
.format(rmsd.max(), rmsd.mean()))
if prefix == '':
prefix = pdb.getTitle() + '_aligned'
outfn = prefix + '.pdb'
LOGGER.info('Writing file: ' + outfn)
prody.writePDB(outfn, pdb)
else:
reffn = args.pop(0)
LOGGER.info('Aligning structures onto: ' + reffn)
ref = prody.parsePDB(reffn)
for arg in args:
if arg == reffn:
continue
if '_aligned.pdb' in arg:
continue
pdb = prody.parsePDB(arg)
if prody.matchAlign(pdb, ref):
outfn = pdb.getTitle() + '_aligned.pdb'
LOGGER.info('Writing file: ' + outfn)
prody.writePDB(outfn, pdb)
else:
LOGGER.warning('Failed to align ' + arg)
def test(pdb='2nwl-mem.pdb', blk='2nwl.blk'):
from prody import parsePDB
from numpy import zeros
pdb = parsePDB(pdb, subset='ca')
pdb.setData('block', zeros(len(pdb), int))
with open(blk) as inp:
for line in inp:
if line.startswith('BLOCK'):
_, b, n1, c1, r1, n2, c2, r2 = line.split()
sel = pdb.select('chain {} and resnum {} to {}'
.format(c1, r1, r2))
if sel:
sel.setData('block', int(b))
pdb.setBetas(pdb.getData('block'))
from prody import writePDB
writePDB('pdb2gb1_truncated.pdb', pdb)
rtb = RTB('2nwl')
rtb.buildHessian(pdb, pdb.getData('block'))
return rtb
def preprocess_pdb(args):
pdb_file = args[1]
output = "./" + args[2]+"/"+args[2]
create_directory("./" + args[2])
cluster_frames = get_frame_numbers(args)
pdb = prody.parsePDB(pdb_file)
# Get a copy of the pdb coords
input_coordsets = numpy.array(pdb.getCoordsets()[cluster_frames])
# Empty pdb
pdb.delCoordset(range(pdb.numCoordsets()))
# Build another pdb to store it
input_pdb = prody.parsePDB(pdb_file)
input_pdb.delCoordset(range(input_pdb.numCoordsets()))
# And add the chosen coordsets
for i in range(len(cluster_frames)):
input_pdb.addCoordset(input_coordsets[i])
prody.writePDB(output+"_ini.pdb", input_pdb)
print_matrix(input_pdb.select("name CA").getCoordsets(), output)
return pdb, input_coordsets, cluster_frames, output
def PCA_Analysis(predID, input_pdbs, tmp_mut_pdb):
# http://prody.csb.pitt.edu/tutorials/ensemble_analysis/xray_calculations.html#calculations
ref_structure = prody.parsePDB(tmp_mut_pdb)
ref_selection = ref_structure.select('calpha')
for ref_chain in ref_structure.getHierView():
prody.startLogfile('%s_%s_log'%(predID, ref_chain.getChid()))
ensemble = prody.PDBEnsemble('%s_%s_ensemble'%(predID, ref_chain.getChid()))
ensemble.setAtoms(ref_chain.select('calpha'))
ensemble.setCoords(ref_chain.select('calpha'))
print 'Generating Ensemble...'
for input_pdb in input_pdbs:
structure = prody.parsePDB(input_pdb, subset='calpha')
mappings = prody.mapOntoChain(structure, ref_chain)
atommap = mappings[0][0]
ensemble.addCoordset(atommap, weights=atommap.getFlags('mapped'))
ensemble.iterpose()
prody.closeLogfile('%s_%s_log'%(predID, ref_chain.getChid()))
prody.writePDB('%s_%s_ensemble.pdb'%(predID, ref_chain.getChid()), ensemble)
pca = prody.PCA('%s_%s_pca'%(predID, ref_chain.getChid()))
pca.buildCovariance(ensemble)
pca.calcModes()
print pca.getEigvecs()
def calc_sasa(pdb, sel='protein', per_res=False, path_dssp='C:\Python27\Scripts\dssp-2.0.4-win32.exe'):
import os as os
import tempfile as tempfile
import shutil as shutil
import prody as prody
import numpy as np
dssp_in = 'dssp.in'
prevdir = os.getcwd()
with tempfile.TemporaryDirectory() as tmpdir:
os.chdir(os.path.expanduser(tmpdir))
prody.writePDB(dssp_in+'.pdb', pdb.select(sel))
pdb_dssp = prody.parsePDB(dssp_in+'.pdb')
execDSSP(dssp_in+'.pdb', dssp=path_dssp)
prody.parseDSSP(dssp_in+'.dssp', pdb_dssp)
if per_res:
return(pdb_dssp._data['dssp_acc'])
else:
return(np.sum(pdb_dssp._data['dssp_acc']))
os.chdir(prevdir)
def test(pdb="2nwl-mem.pdb", blk="2nwl.blk"):
from prody import parsePDB
from numpy import zeros, linalg
pdb = parsePDB(pdb, subset="ca")
pdb.setData("block", zeros(len(pdb), int))
with open(blk) as inp:
for line in inp:
if line.startswith("BLOCK"):
_, b, n1, c1, r1, n2, c2, r2 = line.split()
sel = pdb.select("chain {} and resnum {} to {}".format(c1, r1, r2))
if sel:
sel.setData("block", int(b))
pdb.setBetas(pdb.getData("block"))
coords = pdb.getCoords()
blocks = pdb.getBetas()
from prody import writePDB
writePDB("pdb2gb1_truncated.pdb", pdb)
rtb = RTB("2nwl")
rtb.buildHessian(coords, blocks, scale=64)
# rtb.calcModes()
return rtb
def prody_select(selstr, *pdbs, **kwargs):
"""Write selected atoms from a PDB file in PDB format.
:arg selstr: atom selection string, see :ref:`selections`
:arg pdbs: :term:`PDB` identifier(s) or filename(s)
:arg output: output filename, default is :file:`pdb_selected.pdb`
:arg prefix: prefix for output file, default is PDB filename
:arg suffix: output filename suffix, default is :file:`_selected`"""
from os.path import isfile
from prody import LOGGER, parsePDB, writePDB
#selstr = kwargs.get('selstr')
if not pdbs:
raise ValueError('pdb argument must be provided')
if ((isfile(selstr) or len(selstr) == 4 and selstr[0].isdigit()) and
len(pdbs) == 1 and not isfile(pdbs[0])):
pdbs, selstr = selstr, pdbs[0]
LOGGER.warn('The order of selstr and pdb arguments have switched '
'to support multiple files, old order will be supported '
'until v1.4.')
pdbs = [pdbs]
prefix = kwargs.get('prefix', None)
suffix = kwargs.get('suffix', '_selected')
output = kwargs.get('output', None)
for pdb in pdbs:
pdb = parsePDB(pdb)
pdbselect = pdb.select(selstr)
if pdbselect is None:
LOGGER.warn('Selection {0:s} did not match any atoms.'
.format(repr(selstr)))
return
LOGGER.info('Selection {0:s} matched {1:d} atoms.'
.format(repr(selstr), len(pdbselect)))
outname = output or ((prefix or pdb.getTitle()) + suffix)
LOGGER.info('Selection is written into: ' +
writePDB(outname, pdbselect))
import prody
import numpy
import sys
import os.path
pdb_data = prody.parsePDB(sys.argv[1])
pdb_name = os.path.basename(sys.argv[1])
pdb_trajectory = prody.PDBEnsemble("aligned_CA")
prot = pdb_data.select("name CA not resname CA")
pdb_trajectory.setCoords(pdb_data.getCoordsets()[0])
pdb_trajectory.addCoordset(pdb_data.getCoordsets())
pdb_trajectory.setAtoms(prot)
pdb_trajectory.superpose()
prody.writePDB(pdb_name+".aligned_CA.pdb", pdb_trajectory)
with file(pdb_name+"aligned_CA.coords", 'w') as outfile:
outfile.write("%d %d %d\n"%pdb_trajectory.getCoordsets().shape)
for coordset in pdb_trajectory.getCoordsets():
numpy.savetxt(outfile, coordset)
lig = pdb_data.select("resname BEN not name H1 H2 H3 H4 H5 H6 H7 HN1 HN2")
pdb_trajectory.setAtoms(lig)
rmsds = pdb_trajectory.getRMSDs()
prody.writePDB(pdb_name+".aligned_BEN.pdb", pdb_trajectory)
with file(pdb_name+".aligned_BEN.coords", 'w') as outfile:
outfile.write("%d %d %d\n"%pdb_trajectory.getCoordsets().shape)
for coordset in pdb_trajectory.getCoordsets():
numpy.savetxt(outfile, coordset)
def corepagecalculation(pdbfilename, selatom, noma1, nummodes, gamcut, cut1, gam2, cut2, showresults, smodes, snmd, smodel, scollec, massnomass, sample1, modeens, confens, rmsdens, traverse1, modetra, steptra, rmsdtra, modelnumber, caanm, cagnm, nohanm, nohgnm, allanm, allgnm, bbanm, bbgnm, scanm, scgnm, nmdfolder, modesfolder, collectivityfolder, modelnewname, nmdnewname, modesnewname, modesendname, collectivitynewname, collectivityendname, samplenewname, traversenewname, crosscorr=0, corrfolder='', corrname='', corrend='', compmode01='7', compmode02='15', sqflucts=0, sqfluctsfolder='', sqfluctsname='', sqfluctsend='', separatevar1='0', temfac=0, temfacfolder='', temfacname='', temfacend='', fracovar=0, fraconame='', fracoend='', ovlap=0, ovlapfold='', ovlapname='', ovlapend='', ovlaptab=0, ovlaptabname='', ovlaptabend='', comppdbfilename=''):
# modelnumber
import prody
import time
import os
import Tkinter
root=Tkinter.Tk()
root.title('Info')
onlypage=Tkinter.Frame(root)
onlypage.pack(side='top')
Tkinter.Label(onlypage,text='File: '+pdbfilename).grid(row=0,column=0,sticky='w')
Tkinter.Label(onlypage,text='Atoms: '+selatom).grid(row=1,column=0,sticky='w')
Tkinter.Label(onlypage,text='Analysis: '+noma1).grid(row=2,column=0,sticky='w')
path=os.path.join(os.path.expanduser('~'),'.noma/')
fin = open(path+'savefile.txt','r')
global savedfile
savedfile=fin.readlines()
fin.close()
i=0
a=len(savedfile)
while i<a:
savedfile[i]=savedfile[i][:-1]
i+=1
if gamcut=='0':
Tkinter.Label(onlypage,text='Gamma: r^'+savedfile[91]).grid(row=3,column=0,sticky='w')
Tkinter.Label(onlypage,text='Cutoff: '+cut1).grid(row=4,column=0,sticky='w')
elif gamcut=='1':
Tkinter.Label(onlypage,text='Gamma: '+gam2).grid(row=3,column=0,sticky='w')
Tkinter.Label(onlypage,text='Cutoff: '+cut2).grid(row=4,column=0,sticky='w')
find = 0 #
while find < len(pdbfilename): #
if pdbfilename[-(find+1):-find] == '/': #
bgn = len(pdbfilename)-find #
break #
else: # helps in the
find +=1 # saving of files
try: #
float(bgn) #
except (NameError): #
bgn = 0 #
find = 0 #
while bgn+find<len(pdbfilename): #
if pdbfilename[bgn+find:bgn+find+1] == '.': #
end = len(pdbfilename)-(bgn+find) #
break #
else: #
find +=1 #
try: #
name = pdbfilename[bgn:-end] #
except (NameError): #
name = pdbfilename[bgn:len(pdbfilename)] # name of the file
bgn = pdbfilename[:bgn] # path for file
mytimeis = time.asctime(time.localtime(time.time()))
start = time.time()
try:
p38 = prody.parsePDB(pdbfilename,model=int(modelnumber))
except:
import tkMessageBox
tkMessageBox.askokcancel("File Error","""This is not the correct path or name. Try entering /some/path/nameoffile.pdb
If you need help finding the path, open a new terminal and enter:
find -name 'filename.pdb' use the output as the pdb input
If this doesn't work, make sure the file is in PDB format.""")
p38 = prody.parsePDB(pdbfilename)
print 'Submitted: '+pdbfilename+' at '+mytimeis
Tkinter.Label(onlypage,text='Submitted at: '+mytimeis).grid(row=5,column=0,sticky='w')
root.update()
if selatom == "C-alpha" and noma1 == "Gaussian Normal Mode":
folder = cagnm+'/'
pro = p38.select('protein and name CA') # selects only carbon alpahs
elif selatom == "C-alpha" and noma1 == "Anisotropic Normal Mode":
folder = caanm+'/'
pro = p38.select('protein and name CA')
elif selatom == "Heavy" and noma1 == "Gaussian Normal Mode":
folder = nohgnm+'/'
pro = p38.select('protein and not name "[1-9]?H.*"') # gets rid of all Hydrogens
elif selatom == "Heavy" and noma1 == "Anisotropic Normal Mode":
folder = nohanm+'/'
pro = p38.select('protein and not name "[1-9]?H.*"')
elif selatom == "All" and noma1 == "Gaussian Normal Mode":
folder = allgnm+'/'
pro = p38.select('protein')
elif selatom == "All" and noma1 == "Anisotropic Normal Mode":
folder = allanm+'/'
pro = p38.select('protein')
elif selatom == "Backbone" and noma1 == "Gaussian Normal Mode":
folder = bbgnm+'/'
pro = p38.select('protein and name CA C O N H') # selects backbone
elif selatom == "Backbone" and noma1 == "Anisotropic Normal Mode":
folder = bbanm+'/'
pro = p38.select('protein and name CA C O N H') # selects backbone
elif selatom == "Sidechain" and noma1 == "Gaussian Normal Mode":
folder = scgnm+'/'
pro = p38.select('protein and not name CA C O N H') # selects sidechain
elif selatom == "Sidechain" and noma1 == "Anisotropic Normal Mode":
folder = scanm+'/'
pro = p38.select('protein and not name CA C O N H') # selects sidechain
try: #
open(bgn+folder) # creates the folders
except (IOError): # where the files will
try: # be saved only if they
os.makedirs(bgn+folder) # are not there
except (OSError): #
mer = 0 #
if noma1 == "Gaussian Normal Mode":
print 'Building the Kirchhoff matrix'
Tkinter.Label(onlypage,text='Building Kirchhoff').grid(row=6,column=0,sticky='w')
root.update()
anm = prody.GNM(name)###
if gamcut=='0':
anm.buildKirchhoff(pro,cutoff=float(cut1),gamma=gammaDistanceDependent)###
anm.setKirchhoff(anm.getKirchhoff())
elif gamcut=='1':
anm.buildKirchhoff(pro,cutoff=float(cut2),gamma=float(gam2))###
brat = 2
elif noma1 == "Anisotropic Normal Mode":
print 'Building the Hessian matrix'
Tkinter.Label(onlypage,text='Building Hessian').grid(row=6,column=0,sticky='w')
root.update()
anm = prody.ANM(name)###
if gamcut=='0':
anm.buildHessian(pro,cutoff=float(cut1),gamma=gammaDistanceDependent)###
anm.setHessian(anm.getHessian())###
elif gamcut=='1':
anm.buildHessian(pro,cutoff=float(cut2),gamma=float(gam2))###
brat = 7
print 'Calculating modes'
Tkinter.Label(onlypage,text='Calculating modes').grid(row=7,column=0,sticky='w')
root.update()
anm.calcModes(int(nummodes),zeros = True)###
numatom=anm.numAtoms()###
eigval=anm.getEigvals()###
atomname=pro.getNames()###
if smodel==1:
if brat==2:
modelfilename=bgn+folder+name+modelnewname+'.gnm.npz'
elif brat==7:
modelfilename=bgn+folder+name+modelnewname+'.anm.npz'
print 'Saving Model'
Tkinter.Label(onlypage,text='Saving Model').grid(row=8,column=0,sticky='w')
root.update()
try:
prody.saveModel(anm,bgn+folder+name+modelnewname,True)###
except:
print 'Matrix not saved due to size'
Tkinter.Label(onlypage,text='Matrix not saved').grid(row=8,column=0,sticky='w')
root.update()
prody.saveModel(anm,bgn+folder+name+modelnewname)###
if snmd==1:
print 'Saving NMD'
Tkinter.Label(onlypage,text='Saving NMD').grid(row=9,column=0,sticky='w')
root.update()
try: #
os.makedirs(bgn+folder+nmdfolder+'/') #
except (OSError): #
mer = 0 #
prody.writeNMD(bgn+folder+nmdfolder+'/'+name+nmdnewname+'.nmd',anm[:len(eigval)],pro)### # this can be viewed in VMD
if smodes==1:
print 'Saving Modes'
Tkinter.Label(onlypage,text='Saving Modes').grid(row=10,column=0,sticky='w')
root.update()
try: #
os.makedirs(bgn+folder+modesfolder+'/') #
except (OSError): #
mer = 0 #
modefile = bgn+folder+modesfolder+'/'+name+modesnewname+'.'+modesendname
fout = open(modefile,'w')
mer = 0
while mer< len(eigval):
slowest_mode = anm[mer]###
r = slowest_mode.getEigvec()###
p = slowest_mode.getEigval()###
tq = 0
tt = 0
ttt = 1
tttt = 2
fout.write('MODE {0:3d} {1:15e}'.format(mer+1,p))
fout.write("""
-------------------------------------------------
""")
if noma1 == "Gaussian Normal Mode":
while tq < numatom:
fout.write("""{0:4s}{1:15e}
""".format(atomname[tq],r[tq]))
tq +=1
elif noma1 == "Anisotropic Normal Mode":
while tt < numatom*3:
fout.write("""{0:4s}{1:15e}{2:15e}{3:15e}
""".format(atomname[tq],r[tt],r[ttt],r[tttt]))
tq+=1
tt +=3
ttt+=3
tttt+=3
mer +=1
fout.close()
if showresults=='1':
os.system('/usr/bin/gnome-open '+modefile)
if scollec==1:
print 'Saving collectivity'
Tkinter.Label(onlypage,text='Saving collectivity').grid(row=11,column=0,sticky='w')
root.update()
try: #
os.makedirs(bgn+folder+collectivityfolder+'/') #
except (OSError): #
mer = 0 #
mer = 0
xx = [0]*(numatom) # sets the array to zero and other initial conditions
i = 0
aa = 0
no = 0
var3 = 0
sss = [0]*(len(eigval))
while mer< len(eigval):
slowest_mode = anm[mer]###
r = slowest_mode.getEigvec()###
p = slowest_mode.getEigval()###
a = 0
tt = 0
ttt = 1
tttt = 2
while a < numatom:
atom = atomname[a]
mass = 0
while mass < 2:
if atom[mass] == "N": # all nitrogen
m = 14.0067
break
elif atom[mass] == 'H': # all hydrogen
m = 1.00794
break
elif atom[mass] == "C" : # all carbon
m = 12.0107
break
elif atom[mass] == "O" : # all oxygen
m = 15.9994
break
elif atom[mass] == 'S': # all sulfur
m = 32.065
break
elif atom[mass] == 'P' : # all phosphorus
m = 30.973762
break
else:
if mass == 0:
mass +=1
try:
atom[mass]
except (IndexError):
m = 1
if no == 0:
print 'Enter atom '+atom+' in to the system. Its mass was set to 1 in this simulation.'
no +=1
break
else:
m = 1
if no == 0:
print 'Enter atom '+atom+' in to the system. Its mass was set to 1 in this simulation'
no +=1
break
if len(r)/numatom == 3:
xx[i] = (r[tt]**2 + r[ttt]**2 + r[tttt]**2)/m
i +=1
tt +=3
ttt+=3
tttt+=3
else:
xx[i] = (r[tt]**2)/m
i +=1
tt +=1
a +=1
var3 = 0
j = 0
loop = 1
while loop == 1:
if sum(xx) == 0: # need this because you can't divide by 0
loop = 0
elif j <(numatom):
var1 = xx[j]/sum(xx)
if var1 == 0:
var2 = 0
elif var1 != 0:
from math import log # this means natural log
var2 = var1* log(var1)
var3 += var2
j +=1
else:
from math import exp
k = exp(-var3)/numatom
sss[aa] = k, aa+1
aa +=1
mer +=1
loop = 0
i = 0
xx = [0]*(numatom) # goes through all this until the big loop is done
a = 0
k=[0]*(len(eigval))
while a < len(eigval):
k[a]=prody.calcCollectivity(anm[a]),a+1
a +=1
collectivefile = bgn+folder+collectivityfolder+'/'+name+collectivitynewname+'.'+collectivityendname
fout = open(collectivefile,'w')
if massnomass=='0':
fout.write('MODE COLLECTIVITY(mass)')
fout.write("""
---------------------------
""")
for h in sorted(sss,reverse=True):
fout.write(str(h)[-3:-1]+' '+str(h)[1:19]+"""
""")
fout.write("""
MODE COLLECTIVITY(without mass)""")
fout.write("""
---------------------------
""")
for hh in sorted(k,reverse=True):
fout.write(str(hh)[-3:-1]+' '+str(hh)[1:19]+"""
""")
elif massnomass=='1':
fout.write('MODE COLLECTIVITY(without mass)')
fout.write("""
---------------------------
""")
for hh in sorted(k,reverse=True):
fout.write(str(hh)[-3:-1]+' '+str(hh)[1:19]+"""
""")
fout.write("""
MODE COLLECTIVITY(mass)""")
fout.write("""
---------------------------
""")
for h in sorted(sss,reverse=True):
fout.write(str(h)[-3:-1]+' '+str(h)[1:19]+"""
""")
fout.close()
if showresults=='1':
os.system('/usr/bin/gnome-open '+collectivefile)
fin = open(collectivefile,'r')
lst = fin.readlines()
hi0 = 2
looop = 1
prut=0
secoll=0
thicoll=0
while looop == 1:
fine = lst[hi0]
if int(fine[0:2]) >= brat:
if prut==0:
prut=fine[0:2]
elif secoll==0:
secoll=fine[0:2]
elif thicoll==0:
thicoll=fine[0:2]
else:
foucoll=fine[0:2]
looop = 0
else:
hi0 +=1
mostcollective= "Mode "+prut+" is the most collective."
Tkinter.Label(onlypage,text='Mode '+prut+' is the most collective').grid(row=12,column=0,sticky='w')
root.update()
print mostcollective
fin.close()
if sample1 == 1:
print 'Saving sample file'
Tkinter.Label(onlypage,text='Saving sample file').grid(row=13,column=0,sticky='w')
root.update()
a = modeens+' '
b = [0]*(len(a)+1)
i = 0
j = 0
b1 = 0
while i < len(a):
if a[i:i+1] ==' ' or a[i:i+1]==',':
try:
b[b1]=int(a[j:i])-1
except:
if '1c' in a[j:i]:
b[b1]=int(prut)-1
elif '2c' in a[j:i]:
b[b1]=int(prut)-1
b1 +=1
b[b1]=int(secoll)-1
elif '3c' in a[j:i]:
b[b1]=int(prut)-1
b1 +=1
b[b1]=int(secoll)-1
b1 +=1
b[b1]=int(thicoll)-1
elif '4c' in a[j:i]:
b[b1]=int(prut)-1
b1 +=1
b[b1]=int(secoll)-1
b1 +=1
b[b1]=int(thicoll)-1
b1+=1
b[b1]=int(foucoll)-1
j = i+1
i +=1
b1 +=1
else:
i +=1
del b[b1:]
ensemble = prody.sampleModes(anm[b],pro, n_confs=int(confens), rmsd =float(rmsdens))
p38ens=pro.copy()
p38ens.delCoordset(0)
p38ens.addCoordset(ensemble.getCoordsets())
prody.writePDB(bgn+folder+name+samplenewname+'.pdb',p38ens)
if traverse1 ==1:
print 'Saving traverse file'
Tkinter.Label(onlypage,text='Saving traverse file').grid(row=14,column=0,sticky='w')
root.update()
if modetra=='c':
modefortra=int(prut)-1
else:
modefortra=int(modetra)-1
trajectory=prody.traverseMode(anm[modefortra],pro,n_steps=int(steptra),rmsd=float(rmsdtra))
prody.calcRMSD(trajectory).round(2)
p38traj=pro.copy()
p38traj.delCoordset(0)
p38traj.addCoordset(trajectory.getCoordsets())
prody.writePDB(bgn+folder+name+'_mode'+str(modefortra+1)+traversenewname+'.pdb',p38traj)
if crosscorr==1:
print 'Saving cross correlation'
Tkinter.Label(onlypage,text='Saving cross-correlation').grid(row=15,column=0,sticky='w')
root.update()
try: #
os.makedirs(bgn+folder+corrfolder+'/') #
except (OSError): #
mer = 0
i=int(compmode01)
while i <= int(compmode02):
x=i-1
correlationdataname=bgn+folder+corrfolder+'/'+name+corrname+'_mode'+str(x+1)+'.'+corrend
prody.writeArray(correlationdataname,prody.calcCrossCorr(anm[x]),'%.18e')
print correlationdataname
i+=1
##
if sqflucts==1:
print 'Saving square fluctuation'
Tkinter.Label(onlypage,text='Saving square fluctuation').grid(row=16,column=0,sticky='w')
root.update()
try: #
os.makedirs(bgn+folder+sqfluctsfolder+'/') #
except (OSError): #
mer = 0
i=int(compmode01)
while i < int(compmode02):
yelp = i-1
sqfluctdataname = bgn+folder+sqfluctsfolder+'/'+name+sqfluctsname+'_mode'+str(yelp+1)+'.'+sqfluctsend
fout = open(sqfluctdataname,'w')
if separatevar1=='0':
a = 0
while a < numatom:
fout.write(str(a))
fout.write(""" """)
fout.write(str(prody.calcSqFlucts(anm[yelp])[a]))
fout.write("""
""")
a +=1
elif separatevar1=='1':
a=0
while a <numatom:
firstresnum=int(p38.getResnums()[0:1][0])
origiresnum=int(p38.getResnums()[0:1][0])
while firstresnum<(int(numatom*1.0/p38.numChains())+origiresnum):
fout.write(str(firstresnum))
fout.write('\t')
fout.write(str(prody.calcSqFlucts(anm[yelp])[a]))
fout.write('\n')
a+=1
firstresnum+=1
fout.write('&\n')
fout.close()
print sqfluctdataname
i+=1
if temfac==1:
print 'Saving temperature factors'
Tkinter.Label(onlypage,text='Saving temperature factors').grid(row=17,column=0,sticky='w')
root.update()
try: #
os.makedirs(bgn+folder+temfacfolder+'/') #
except (OSError): #
mer = 0
fin=open(pdbfilename,'r')
d = [None]*len(atomname)
e = 0
for line in fin:
pair = line.split()
if 'ATOM ' in line and e < len(atomname):
if str(pair[2]) == str(atomname[e]):
d[e]=str(pair[1])
e+=1
else:
e+=0
else:
continue
fin.close()
sqf = prody.calcSqFlucts(anm)
x = sqf/((sqf**2).sum()**.5)
y = prody.calcTempFactors(anm,pro)
a = 0
tempfactorsdataname =bgn+folder+temfacfolder+'/'+name+temfacname+'.'+temfacend
fout=open(tempfactorsdataname,'w')
fout.write("""Atom Residue TempFactor TempFactor with exp beta
""")
while a < numatom:
fout.write("""{0:4s} {1:4d} {2:15f} {3:15f}
""".format(d[a],a+1,x[a],y[a]))
a +=1
fout.close()
print tempfactorsdataname
if fracovar==1:
try:
import matplotlib.pyplot as plt
print 'Saving Fraction of Variance'
Tkinter.Label(onlypage,text='Saving Fraction of Variance').grid(row=18,column=0,sticky='w')
root.update()
try: #
os.makedirs(bgn+folder+modesfolder+'/') #
except (OSError): #
mer = 0 #
plt.figure(figsize = (5,4))
prody.showFractVars(anm)
prody.showCumulFractVars(anm)
fracvardataname =bgn+folder+modesfolder+'/'+name+fraconame+'.'+fracoend
plt.savefig(fracvardataname)
print fracvardataname
if showresults=='1':
os.system('/usr/bin/gnome-open '+fracvardataname)
except:
print 'Error: Fraction of Variance'
Tkinter.Label(onlypage,text='Error: Fraction of Variance').grid(row=18,column=0,sticky='w')
root.update()
mer=0
if ovlap==1 or ovlaptab==1:
try:
import matplotlib.pyplot as plt
print 'Saving Overlap'
Tkinter.Label(onlypage,text='Saving Overlap').grid(row=19,column=0,sticky='w')
root.update()
Tkinter.Label(onlypage,text='Comparison: '+comppdbfilename).grid(row=20,column=0,sticky='w')
##
find = 0
while find < len(comppdbfilename):
if comppdbfilename[-(find+1):-find] == '/':
bgn1 = len(comppdbfilename)-find
break
else:
find +=1
try:
float(bgn1)
except (NameError):
bgn1 = 0
find = 0
while bgn1+find<len(comppdbfilename):
if comppdbfilename[bgn1+find:bgn1+find+1] == '.':
end1 = len(comppdbfilename)-(bgn1+find)
break
else:
find +=1
try:
name1 = comppdbfilename[bgn1:-end1]
except (NameError):
name1 = comppdbfilename[bgn1:len(comppdbfilename)]
bgn1 = comppdbfilename[:bgn1]
p381 = prody.parsePDB(comppdbfilename,model=int(modelnumber))
if selatom == "C-alpha" and noma1 == "Gaussian Normal Mode":
pro1 = p381.select('protein and name CA')
elif selatom == "C-alpha" and noma1 == "Anisotropic Normal Mode":
pro1 = p381.select('protein and name CA')
elif selatom == "Heavy" and noma1 == "Gaussian Normal Mode":
pro1 = p381.select('protein and not name "[1-9]?H.*"')
elif selatom == "Heavy" and noma1 == "Anisotropic Normal Mode":
pro1 = p381.select('protein and not name "[1-9]?H.*"')
elif selatom == "All" and noma1 == "Gaussian Normal Mode":
pro1 = p381.select('protein')
elif selatom == "All" and noma1 == "Anisotropic Normal Mode":
pro1 = p381.select('protein')
elif selatom == "Backbone" and noma1 == "Gaussian Normal Mode":
pro1 = p381.select('protein and name CA C O N H')
elif selatom == "Backbone" and noma1 == "Anisotropic Normal Mode":
pro1 = p381.select('protein and name CA C O N H')
elif selatom == "Sidechain" and noma1 == "Gaussian Normal Mode":
pro1 = p381.select('protein and not name CA C O N H')
elif selatom == "Sidechain" and noma1 == "Anisotropic Normal Mode":
pro1 = p381.select('protein and not name CA C O N H')
if noma1 == "Gaussian Normal Mode":
print 'Building the Kirchhoff matrix'
Tkinter.Label(onlypage,text='Building Kirchhoff').grid(row=21,column=0,sticky='w')
root.update()
anm1 = prody.GNM(name1)
if gamcut=='0':
anm1.buildKirchhoff(pro1,cutoff=float(cut1),gamma=gammaDistanceDependent)
anm1.setKirchhoff(anm1.getKirchhoff())
elif gamcut=='1':
anm1.buildKirchhoff(pro1,cutoff=float(cut2),gamma=float(gam2))
brat = 2
elif noma1 == "Anisotropic Normal Mode":
print 'Building the Hessian matrix'
Tkinter.Label(onlypage,text='Building Hessian').grid(row=21,column=0,sticky='w')
root.update()
anm1 = prody.ANM(name1)
if gamcut=='0':
anm1.buildHessian(pro1,cutoff=float(cut1),gamma=gammaDistanceDependent)
anm1.setHessian(anm1.getHessian())
elif gamcut=='1':
anm1.buildHessian(pro1,cutoff=float(cut2),gamma=float(gam2))
brat = 7
print 'Calculating modes'
Tkinter.Label(onlypage,text='Calculating modes').grid(row=22,column=0,sticky='w')
root.update()
anm1.calcModes(int(nummodes),zeros = True)
##
try:
os.makedirs(bgn+folder+ovlapfold+'/')
except (OSError):
mer = 0
if ovlap==1:
i=int(compmode01)
while i < int(compmode02):
a = i-1
plt.figure(figsize=(5,4))
prody.showCumulOverlap(anm[a],anm1)
prody.showOverlap(anm[a],anm1)
plt.title('Overlap with Mode '+str(a+1)+' from '+name)
plt.xlabel(name1+' mode index')
overlapname = bgn+folder+ovlapfold+'/'+name+'_'+name1+ovlapname+'_mode'+str(a+1)+'.'+ovlapend
plt.savefig(overlapname)
print overlapname
i+=1
if ovlaptab==1:
plt.figure(figsize=(5,4))
prody.showOverlapTable(anm1,anm)
plt.xlim(int(compmode01)-1,int(compmode02))
plt.ylim(int(compmode01)-1,int(compmode02))
plt.title(name1+' vs '+name+' Overlap')
plt.ylabel(name1)
plt.xlabel(name)
overlapname = bgn+folder+ovlapfold+'/'+name+'_'+name1+ovlaptabname+'.'+ovlaptabend
plt.savefig(overlapname)
print overlapname
except:
mer=0
root.destroy()
mynewtimeis = float(time.time()-start)
if mynewtimeis <= 60.00:
timeittook= "The calculations took %.2f s."%(mynewtimeis)
elif mynewtimeis > 60.00 and mynewtimeis <= 3600.00:
timeittook= "The calculations took %.2f min."%((mynewtimeis/60.00))
else:
timeittook= "The calculations took %.2f hrs."%((mynewtimeis/3600.00))
print timeittook
if smodel==1 and scollec==1:
return (timeittook,modelfilename,str(int(prut)))
elif scollec==1:
return (timeittook,'nofile',str(int(prut)))
elif smodel==1:
return (timeittook,modelfilename,'nocoll')
else:
return (timeittook,'nofile','nocoll')
import prody
import numpy
pdb_data = prody.parsePDB("../Models/prot_plus_ligand_very_different/not_aligned_offset_prot_plus_ligand.pdb")
pdb_trajectory = prody.PDBEnsemble("aligned_CA")
prot = pdb_data.select("name CA not resname CA")
pdb_trajectory.setCoords(pdb_data.getCoordsets()[0])
pdb_trajectory.addCoordset(pdb_data.getCoordsets())
pdb_trajectory.setAtoms(prot)
pdb_trajectory.superpose()
prody.writePDB("prot_plus_ligand_similar.aligned_CA.pdb", pdb_trajectory)
with file("prot_plus_ligand_similar.aligned_CA.coords", 'w') as outfile:
outfile.write("%d %d %d\n"%pdb_trajectory.getCoordsets().shape)
for coordset in pdb_trajectory.getCoordsets():
numpy.savetxt(outfile, coordset)
lig = pdb_data.select("resname BEN not element H")
pdb_trajectory.setAtoms(lig)
rmsds = pdb_trajectory.getRMSDs()
prody.writePDB("prot_plus_ligand_similar.aligned_BEN.pdb", pdb_trajectory)
with file("prot_plus_ligand_similar.aligned_BEN.coords", 'w') as outfile:
outfile.write("%d %d %d\n"%pdb_trajectory.getCoordsets().shape)
for coordset in pdb_trajectory.getCoordsets():
numpy.savetxt(outfile, coordset)
print rmsds
numpy.savetxt("prot_plus_ligand_similar.aligned_BEN.rmsd", rmsds)
pdb, pdb_path = tools.get_pdb_from_remote_or_db(alignment["pdb"]["id"],
parameters["pdb_preparation"]["load_selection"],
structure_db_path)
main_chains = choose_main_chains(pdb)
curated_pdb, ligand = curate_struct(pdb,
main_chains,
alignment,
parameters["pdb_preparation"])
failed_filters = structure_filter.must_be_filtered(curated_pdb)
if len(failed_filters) == 0 or alignment["pdb"]["id"] in id_exceptions:
alignment["rejected"] = False
tmp_path = os.path.join(parameters["global"]["curated_structure_database"],
os.path.basename(pdb_path))
prody.writePDB(tmp_path, curated_pdb)
else:
alignment["rejected"] = True
log.write("\t - PDB %s is not going to be stored because it has failed this filters: %s.\n"%(alignment["pdb"]["id"],
str(failed_filters)))
if structure_db_path is not None: # We do not want to delete our database!
os.remove(pdb_path)
# Blast DB creation
BlastpCommands.create_database_from_alignments(filtered_alignments,
parameters["blast_database_creation"])
tools.save_json(filtered_alignments,
parameters["blastp"]["alignments_file"])
log.write(datetime.now().strftime("Finished on %A, %d. %B %Y %I:%M%p\n"))
import prody
import numpy
pdb_data = prody.parsePDB("../Models/prot_stretching/stretching_trajectory_offset_ligand.pdb")
pdb_trajectory = prody.PDBEnsemble("iterposed_CA")
# Write the initial coordsets
prot = pdb_data.select("name CA")
prody.writePDB("stretching_trajectory_offset_ligand.iterposed_all.pdb", prot)
with file("stretching_trajectory_offset_ligand.initial_CA.coords", 'w') as outfile:
outfile.write("%d %d %d\n"%prot.getCoordsets().shape)
for coordset in prot.getCoordsets():
numpy.savetxt(outfile, coordset)
# We only want to work with CAs. If we use the 'all coordinates+atom selection" trick
# Prody will still use all coordinates for iterative superposition
pdb_trajectory.setCoords(prot.getCoordsets()[0])
pdb_trajectory.addCoordset(prot.getCoordsets())
pdb_trajectory.setAtoms(prot)
pdb_trajectory.iterpose()
prody.writePDB("stretching_trajectory_offset_ligand.iterposed_CA.pdb", pdb_trajectory)
with file("stretching_trajectory_offset_ligand.iterposed_CA.coords", 'w') as outfile:
outfile.write("%d %d %d\n"%pdb_trajectory.getCoordsets().shape)
for coordset in pdb_trajectory.getCoordsets():
numpy.savetxt(outfile, coordset)
import sys
import prody
import numpy
import os.path
pdb_data = prody.parsePDB(sys.argv[1])
pdb_trajectory = prody.PDBEnsemble("aligned_CA")
pdb_name = os.path.splitext(sys.argv[1])[0]
prot = pdb_data.select("name CA not resname CA")
pdb_trajectory.setAtoms(prot)
pdb_trajectory.addCoordset(prot.getCoordsets())
pdb_trajectory.setCoords(prot.getCoordsets()[0])
prody.writePDB(pdb_name+"_CA.pdb", pdb_trajectory)
with file(pdb_name+".CA.coords", 'w') as outfile:
outfile.write("%d %d %d\n"%pdb_trajectory.getCoordsets().shape)
for coordset in pdb_trajectory.getCoordsets():
numpy.savetxt(outfile, coordset)
pdb_trajectory = prody.PDBEnsemble("")
lig = pdb_data.select("resname BEN not name H1 H2 H3 H4 H5 H6 H7 HN1 HN2 ")
pdb_trajectory.setAtoms(lig)
pdb_trajectory.addCoordset(lig.getCoordsets())
pdb_trajectory.setCoords(lig.getCoordsets()[0])
prody.writePDB(pdb_name+"_ligand.pdb", pdb_trajectory)
with file(pdb_name+".ligand.coords", 'w') as outfile:
outfile.write("%d %d %d\n"%pdb_trajectory.getCoordsets().shape)
for coordset in pdb_trajectory.getCoordsets():
numpy.savetxt(outfile, coordset)