def _try_as_cif(self):
# XXX hack to avoid choking on CCP4 maps and images
file_ext = os.path.splitext(self.file_name)[1]
assert (not file_ext in [".ccp4", ".img", ".osc", ".mccd"])
import iotbx.cif
from iotbx.reflection_file_reader import any_reflection_file
cif_file = any_reflection_file(str(self.file_name))
if cif_file.file_type() is not None:
self._file_object = cif_file
self._file_type = "hkl"
else:
from iotbx.pdb.mmcif import cif_input
from iotbx.pdb.hierarchy import input_hierarchy_pair
try:
cif_in = cif_input(file_name=self.file_name)
self._file_object = input_hierarchy_pair(
cif_in, cif_in.hierarchy)
self._file_type = "pdb"
except Exception as e:
if (str(e).startswith("Space group is incompatible")
or str(e).startswith("The space group")):
raise
else:
self._file_object = iotbx.cif.reader(
file_path=self.file_name, strict=False)
self._file_type = "cif"
def old_run(args, out=None):
import iotbx.phil
if (out is None):
out = sys.stdout
cmdline = iotbx.phil.process_command_line_with_files(
args=args,
master_phil=master_phil,
pdb_file_def="input.pdb_file",
seq_file_def="input.seq_file")
params = cmdline.work.extract()
cmdline.work.show()
if params.output.cif_file is None:
params.output.cif_file = os.path.splitext(
params.input.pdb_file)[0] + ".deposit.cif"
model_vs_sequence.validate_params(params)
pdb_input = mmcif.cif_input(file_name=params.input.pdb_file)
pdb_hierarchy = pdb_input.construct_hierarchy()
cif_model = pdb_input.cif_model
cif_block = pdb_input.cif_block
seq_in = any_file(params.input.seq_file, force_type="seq")
seq_in.check_file_type("seq")
sequences = seq_in.file_object
cif_block = pdb_hierarchy.as_cif_block_with_sequence(
sequences,
crystal_symmetry=pdb_input.crystal_symmetry(),
alignment_params=params)
block_name = cif_model.keys()[0]
def float_or_none(string):
try:
return float(string)
except TypeError:
return None
d_min_from_cif = float_or_none(cif_block.get('_refine.ls_d_res_high'))
d_max_from_cif = float_or_none(cif_block.get('_refine.ls_d_res_low'))
# XXX maybe the values from the CIF (i.e. those actually used in the refinement)
# should override the input params?
if params.high_resolution is not None:
params.high_resolution = d_min_from_cif
if params.low_resolution is not None:
params.low_resolution = d_max_from_cif
if params.input.unmerged_data is not None:
from mmtbx.command_line import cc_star
result = cc_star.run(params=params.input, out=out)
cif_block.update(result.as_cif_block())
cif_model[block_name].update(cif_block)
cif_model[block_name].sort(key=category_sort_function)
print >> out, "Writing updated CIF file:"
print >> out, " " + params.output.cif_file
with open(params.output.cif_file, "wb") as f:
print >> f, cif_model
return
def run(args, out=None):
import iotbx.phil
if (out is None) :
out = sys.stdout
cmdline = iotbx.phil.process_command_line_with_files(
args=args,
master_phil=master_phil,
pdb_file_def="input.pdb_file",
seq_file_def="input.seq_file"
)
params = cmdline.work.extract()
cmdline.work.show()
if params.output.cif_file is None:
params.output.cif_file = os.path.splitext(params.input.pdb_file)[0] + ".deposit.cif"
model_vs_sequence.validate_params(params)
pdb_input = mmcif.cif_input(file_name=params.input.pdb_file)
pdb_hierarchy = pdb_input.construct_hierarchy()
cif_model = pdb_input.cif_model
cif_block = pdb_input.cif_block
seq_in = any_file(params.input.seq_file, force_type="seq")
seq_in.check_file_type("seq")
sequences = seq_in.file_object
cif_block = pdb_hierarchy.as_cif_block_with_sequence(
sequences, crystal_symmetry=pdb_input.crystal_symmetry(),
alignment_params=params)
block_name = cif_model.keys()[0]
def float_or_none(string):
try: return float(string)
except TypeError: return None
d_min_from_cif = float_or_none(cif_block.get('_refine.ls_d_res_high'))
d_max_from_cif = float_or_none(cif_block.get('_refine.ls_d_res_low'))
# XXX maybe the values from the CIF (i.e. those actually used in the refinement)
# should override the input params?
if params.high_resolution is not None:
params.high_resolution = d_min_from_cif
if params.low_resolution is not None:
params.low_resolution = d_max_from_cif
if params.input.unmerged_data is not None:
from mmtbx.command_line import cc_star
result = cc_star.run(params=params.input, out=out)
cif_block.update(result.as_cif_block())
cif_model[block_name].update(cif_block)
cif_model[block_name].sort(key=category_sort_function)
print >> out, "Writing updated CIF file:"
print >> out, " " + params.output.cif_file
with open(params.output.cif_file, "wb") as f:
print >> f, cif_model
return
def exercise_extract_header_misc():
cif_file = libtbx.env.find_in_repositories(
relative_path="phenix_regression/pdb/3orl.cif", test=os.path.isfile)
if (cif_file is None):
return
cif_in = mmcif.cif_input(file_name=cif_file)
assert (cif_in.file_type() == "mmcif")
wavelength = cif_in.extract_wavelength()
assert (approx_equal(wavelength, 1.8927))
exptl_method = cif_in.get_experiment_type()
assert (exptl_method == 'X-RAY DIFFRACTION')
r_rfree_sigma = cif_in.get_r_rfree_sigma(cif_file)
r_rfree_sigma.show()
def exercise_extract_header_misc () :
cif_file = libtbx.env.find_in_repositories(
relative_path="phenix_regression/pdb/3orl.cif",
test=os.path.isfile)
if (cif_file is None) :
return
cif_in = mmcif.cif_input(file_name=cif_file)
assert (cif_in.file_type() == "mmcif")
wavelength = cif_in.extract_wavelength()
assert (approx_equal(wavelength, 1.8927))
exptl_method = cif_in.get_experiment_type()
assert (exptl_method == 'X-RAY DIFFRACTION')
r_rfree_sigma = cif_in.get_r_rfree_sigma(cif_file)
r_rfree_sigma.show()
def _try_as_cif (self) :
# XXX hack to avoid choking on CCP4 maps and images
file_ext = os.path.splitext(self.file_name)[1]
assert (not file_ext in [".ccp4", ".img", ".osc", ".mccd"])
import iotbx.cif
from iotbx.reflection_file_reader import any_reflection_file
cif_file = any_reflection_file(str(self.file_name))
if cif_file.file_type() is not None:
self._file_object = cif_file
self._file_type = "hkl"
else:
from iotbx.pdb.mmcif import cif_input
from iotbx.pdb.hierarchy import input_hierarchy_pair
try:
cif_in = cif_input(file_name=self.file_name)
self._file_object = input_hierarchy_pair(cif_in, cif_in.hierarchy)
self._file_type = "pdb"
except Exception, e:
self._file_object = iotbx.cif.reader(file_path=self.file_name,
strict=False)
self._file_type = "cif"
def get_pdb_info(cif_file, data_file, dres, matthews=True, chains=True):
"""Get info from PDB of mmCIF file"""
# Get rid of ligands and water so Phenix won't error.
np = 0
na = 0
nmol = 1
sc = 0.55
nchains = 0
res1 = 0.0
d = {}
l = []
# Read in the file
cif_file = convert_unicode(cif_file)
if cif_file[-3:].lower() == 'cif':
root = iotbx_mmcif.cif_input(file_name=cif_file).construct_hierarchy()
else:
root = iotbx_pdb.input(cif_file).construct_hierarchy()
# Go through the chains
for chain in root.models()[0].chains():
# Number of protein residues
np1 = 0
# Number of nucleic acid residues
na1 = 0
# Sometimes Hetatoms are AA with same segid.
if l.count(chain.id) == 0:
l.append(chain.id)
repeat = False
nchains += 1
else:
repeat = True
# Count the number of AA and NA in pdb file.
for rg in chain.residue_groups():
if rg.atoms()[0].parent().resname in iotbx_pdb.common_residue_names_amino_acid:
np1 += 1
if rg.atoms()[0].parent().resname in iotbx_pdb.common_residue_names_rna_dna:
na1 += 1
# Not sure if I get duplicates?
if rg.atoms()[0].parent().resname in \
iotbx_pdb.common_residue_names_ccp4_mon_lib_rna_dna:
na1 += 1
# Limit to 10 chains?!?
if nchains < 10:
# Do not split up PDB if run from cell analysis
if chains and not repeat:
# Save info for each chain.
if np1 or na1:
# Write new pdb files for each chain.
temp = iotbx_pdb.hierarchy.new_hierarchy_from_chain(chain)
# Long was of making sure that user does not have directory named '.pdb' or
# '.cif'
#n = os.path.join(os.path.dirname(cif_file), "%s_%s.pdb" % \
n = os.path.join(os.path.dirname(cif_file), "%s_%s.cif" % \
(os.path.basename(cif_file)[:os.path.basename(cif_file).find('.')], \
chain.id))
#temp.write_pdb_file(file_name=n)
temp.write_mmcif_file(file_name=n)
d[chain.id] = {'file': n,
'NRes': np1+na1,
'MWna': na1*330,
'MWaa': np1*110,
'MW': na1*330+np1*110}
if matthews:
# Run Matthews Calc. on chain
#phaser_return = run_phaser_module((np1, na1, dres, n, data_file))
#phaser_return = run_phaser_module(data_file, (np1, na1, dres, n))
phaser_return = run_phaser_module(data_file=data_file,
ellg=True,
cca=True,
mmcif=n,
dres=dres,
np=np1,
na=na1)
d[chain.id].update({'NMol': phaser_return.get("z", nmol),
'SC': phaser_return.get("solvent_content", sc),
'res': phaser_return.get("target_resolution", res1)})
else:
#res1 = run_phaser_module(n)
phaser_return = run_phaser_module(data_file=data_file,
ellg=True,
mmcif=n)
d[chain.id].update({'res': phaser_return.get("target_resolution", res1)})
"""
d[chain.id] = {'file': n,
'NRes': np1+na1,
'MWna': na1*330,
'MWaa': np1*110,
'MW': na1*330+np1*110,
'NMol': phaser_return.get("z", nmol),
'SC': phaser_return.get("solvent_content", sc),
'res': phaser_return.get("target_resolution", res1)}
"""
# Add up residue count
np += np1
na += na1
d['all'] = {'file': cif_file,
'NRes': np+na,
'MWna': na*330,
'MWaa': np*110,
'MW': na*330+np*110}
# Run on entire PDB
if matthews:
#phaser_return = run_phaser_module((np, na, dres, cif_file, data_file))
#phaser_return = run_phaser_module(data_file, (np, na, dres, cif_file))
phaser_return = run_phaser_module(data_file=data_file,
ellg=True,
cca=True,
mmcif=cif_file,
dres=dres,
np=np,
na=na)
d['all'].update({'NMol': phaser_return.get("z", nmol),
'SC': phaser_return.get("solvent_content", sc),
'res': phaser_return.get("target_resolution", res1)})
else:
#phaser_return = run_phaser_module((np, na, dres, cif_file, data_file))
#phaser_return = run_phaser_module(data_file, (np, na, dres, cif_file))
phaser_return = run_phaser_module(data_file=data_file,
ellg=True,
mmcif=cif_file)
d['all'].update({'res': phaser_return.get("target_resolution", res1)})
"""
d['all'] = {'file': cif_file,
'NRes': np+na,
'MWna': na*330,
'MWaa': np*110,
'MW': na*330+np*110,
'NMol': phaser_return.get("z", nmol),
'SC': phaser_return.get("solvent_content", sc),
'res': phaser_return.get("target_resolution", res1)}
"""
return d
def get_pdb_info(struct_file, data_file, dres, matthews=True, chains=True):
"""Get info from PDB or mmCIF file"""
# Get rid of ligands and water so Phenix won't error.
np = 0
na = 0
nmol = 1
sc = 0.55
nchains = 0
res1 = 0.0
d = {}
l = []
# Read in the file
struct_file = convert_unicode(struct_file)
if struct_file[-3:].lower() == 'cif':
root = iotbx_mmcif.cif_input(
file_name=struct_file).construct_hierarchy()
else:
root = iotbx_pdb.input(struct_file).construct_hierarchy()
# Go through the chains
for chain in root.models()[0].chains():
# Number of protein residues
np1 = 0
# Number of nucleic acid residues
na1 = 0
# Sometimes Hetatoms are AA with same segid.
if l.count(chain.id) == 0:
l.append(chain.id)
repeat = False
nchains += 1
else:
repeat = True
# Count the number of AA and NA in pdb file.
for rg in chain.residue_groups():
if rg.atoms()[0].parent(
).resname in iotbx_pdb.common_residue_names_amino_acid:
np1 += 1
if rg.atoms()[0].parent(
).resname in iotbx_pdb.common_residue_names_rna_dna:
na1 += 1
# Not sure if I get duplicates?
if rg.atoms()[0].parent().resname in \
iotbx_pdb.common_residue_names_ccp4_mon_lib_rna_dna:
na1 += 1
# Limit to 10 chains?!?
if nchains < 10:
# Do not split up PDB if run from cell analysis
if chains and not repeat:
# Save info for each chain.
if np1 or na1:
# Write new pdb files for each chain.
temp = iotbx_pdb.hierarchy.new_hierarchy_from_chain(chain)
# Long was of making sure that user does not have directory named '.pdb' or
# '.cif'
#n = os.path.join(os.path.dirname(struct_file), "%s_%s.pdb" % \
n = os.path.join(os.path.dirname(struct_file), "%s_%s.cif" % \
(os.path.basename(struct_file)[:os.path.basename(struct_file).find('.')], \
chain.id))
#temp.write_pdb_file(file_name=n)
# Write chain as mmCIF file.
temp.write_mmcif_file(file_name=n)
d[chain.id] = {
'file': n,
'NRes': np1 + na1,
'MWna': na1 * 330,
'MWaa': np1 * 110,
'MW': na1 * 330 + np1 * 110
}
if matthews:
# Run Matthews Calc. on chain
#phaser_return = run_phaser_module((np1, na1, dres, n, data_file))
#phaser_return = run_phaser_module(data_file, (np1, na1, dres, n))
phaser_return = run_phaser_module(data_file=data_file,
ellg=True,
cca=True,
struct_file=n,
dres=dres,
np=np1,
na=na1)
d[chain.id].update({
'NMol':
phaser_return.get("z", nmol),
'SC':
phaser_return.get("solvent_content", sc),
'res':
phaser_return.get("target_resolution", res1)
})
else:
#res1 = run_phaser_module(n)
phaser_return = run_phaser_module(data_file=data_file,
ellg=True,
struct_file=n)
d[chain.id].update({
'res':
phaser_return.get("target_resolution", res1)
})
"""
d[chain.id] = {'file': n,
'NRes': np1+na1,
'MWna': na1*330,
'MWaa': np1*110,
'MW': na1*330+np1*110,
'NMol': phaser_return.get("z", nmol),
'SC': phaser_return.get("solvent_content", sc),
'res': phaser_return.get("target_resolution", res1)}
"""
# Add up residue count
np += np1
na += na1
d['all'] = {
'file': struct_file,
'NRes': np + na,
'MWna': na * 330,
'MWaa': np * 110,
'MW': na * 330 + np * 110
}
# Run on entire PDB
if matthews:
#phaser_return = run_phaser_module((np, na, dres, struct_file, data_file))
#phaser_return = run_phaser_module(data_file, (np, na, dres, struct_file))
phaser_return = run_phaser_module(data_file=data_file,
ellg=True,
cca=True,
struct_file=struct_file,
dres=dres,
np=np,
na=na)
d['all'].update({
'NMol': phaser_return.get("z", nmol),
'SC': phaser_return.get("solvent_content", sc),
'res': phaser_return.get("target_resolution", res1)
})
else:
#phaser_return = run_phaser_module((np, na, dres, struct_file, data_file))
#phaser_return = run_phaser_module(data_file, (np, na, dres, struct_file))
# phaser_return = run_phaser_module(data_file=data_file,
# ellg=True,
# struct_file=struct_file)
phaser_return = run_phaser_module(data_file=data_file,
ellg=True,
struct_file=struct_file)
d['all'].update({'res': phaser_return.get("target_resolution", res1)})
"""
d['all'] = {'file': struct_file,
'NRes': np+na,
'MWna': na*330,
'MWaa': np*110,
'MW': na*330+np*110,
'NMol': phaser_return.get("z", nmol),
'SC': phaser_return.get("solvent_content", sc),
'res': phaser_return.get("target_resolution", res1)}
"""
return d