def run(
pdb_filename,
model_completion=True,
keep_alt_loc=False,
skip_validation=False,
calculate_charge=False,
append_to_end_of_model=False,
):
#
# do all *.pdb in a directory
#
if os.path.isdir(pdb_filename):
loop_over_dir(pdb_filename)
return
#
# fetch a PDB code
#
if pdb_filename.find(".pdb") == -1:
if not os.path.exists('%s.pdb' % pdb_filename):
run_fetch_pdb(pdb_filename)
pdb_filename = '%s.pdb' % pdb_filename
if 1: # read file option
ppf = hierarchy_utils.get_processed_pdb(pdb_filename=pdb_filename)
elif 0: # raw records example
pdb_inp = iotbx.pdb.input(pdb_filename)
hierarchy = pdb_inp.construct_hierarchy()
raw_records = []
for atom in hierarchy.atoms():
raw_records.append(atom.format_atom_record())
ppf = hierarchy_utils.get_processed_pdb(raw_records=raw_records)
else: # from hierarchy
pdb_inp = iotbx.pdb.input(pdb_filename)
hierarchy = pdb_inp.construct_hierarchy()
ppf = hierarchy_utils.get_processed_pdb(
pdb_inp=hierarchy.as_pdb_input())
if not skip_validation:
initial_model_statistics = mmtbx.model.statistics.geometry(
pdb_hierarchy=ppf.all_chain_proxies.pdb_hierarchy)
# should use cctbx
if keep_alt_loc: pass
else:
hierarchy = remove_alternative_locations(
ppf.all_chain_proxies.pdb_hierarchy)
ppf.all_chain_proxies.pdb_hierarchy.shift_to_origin(
ppf.all_chain_proxies.pdb_inp.crystal_symmetry())
ppf.all_chain_proxies.pdb_hierarchy.remove_residue_groups_with_atoms_on_special_positions_selective(
ppf.all_chain_proxies.pdb_inp.crystal_symmetry())
if 0:
output = hierarchy_utils.write_hierarchy(
pdb_filename, # uses to get output filename
ppf.all_chain_proxies.pdb_inp,
ppf.all_chain_proxies.pdb_hierarchy,
"temp0")
ppf = hierarchy_utils.get_processed_pdb(pdb_filename=output)
else:
raw_records = hierarchy_utils.get_raw_records(
ppf.all_chain_proxies.pdb_inp,
ppf.all_chain_proxies.pdb_hierarchy,
)
ppf = hierarchy_utils.get_processed_pdb(raw_records=raw_records)
hetero_charges = charges.get_hetero_charges(
ppf.all_chain_proxies.pdb_inp,
ppf.all_chain_proxies.pdb_hierarchy,
)
if not hetero_charges:
# some defaults
hetero_charges = charges.default_ion_charges
inter_residue_bonds = charges.get_inter_residue_bonds(ppf)
#
# extends side chains and add hydrogens
#
if model_completion:
use_capping_hydrogens = False
fname = 'complete'
else:
use_capping_hydrogens = True
fname = 'capping'
#assert 0 # model has H
ppf = completion.complete_pdb_hierarchy(
ppf.all_chain_proxies.pdb_hierarchy,
ppf.geometry_restraints_manager(),
pdb_filename=pdb_filename,
pdb_inp=ppf.all_chain_proxies.pdb_inp,
use_capping_hydrogens=use_capping_hydrogens,
append_to_end_of_model=append_to_end_of_model,
)
if calculate_charge:
new_pdb_filename = hierarchy_utils.write_hierarchy(
pdb_filename, # uses to get output filename
ppf.all_chain_proxies.pdb_inp,
ppf.all_chain_proxies.pdb_hierarchy,
fname)
## need now inter_residue_bonds because of added hydrogens
## maybe update the bond table!!!
ppf = hierarchy_utils.get_processed_pdb(pdb_filename=new_pdb_filename)
inter_residue_bonds = charges.get_inter_residue_bonds(ppf,
verbose=True)
total_charge = charges.calculate_pdb_hierarchy_charge(
ppf.all_chain_proxies.pdb_hierarchy,
hetero_charges=hetero_charges,
inter_residue_bonds=inter_residue_bonds,
)
print "total_charge", total_charge
# Idealize H as riding
params = mmtbx.model.manager.get_default_pdb_interpretation_params()
params.pdb_interpretation.use_neutron_distances = True
h = ppf.all_chain_proxies.pdb_hierarchy
asc = h.atom_selection_cache()
sel = asc.selection("element H or element D")
model = mmtbx.model.manager(
model_input=None,
build_grm=True,
pdb_hierarchy=ppf.all_chain_proxies.pdb_hierarchy,
pdb_interpretation_params=params,
crystal_symmetry=ppf.all_chain_proxies.pdb_inp.crystal_symmetry(),
log=null_out())
model.idealize_h_riding()
model.set_occupancies(value=0, selection=sel)
## after no error getting total charge, write the completed pdb file
hierarchy_utils.write_hierarchy(
pdb_filename, # uses to get output filename
ppf.all_chain_proxies.pdb_inp,
model.get_hierarchy(),
fname)
if not skip_validation:
final_model_statistics = mmtbx.model.statistics.geometry(
pdb_hierarchy=ppf.all_chain_proxies.pdb_hierarchy)
def run(pdb_filename,
model_completion=True,
keep_alt_loc=False,
skip_validation=False,
calculate_charge=False,
):
#
# do all *.pdb in a directory
#
if os.path.isdir(pdb_filename):
loop_over_dir(pdb_filename)
return
#
# fetch a PDB code
#
if pdb_filename.find(".pdb")==-1:
if not os.path.exists('%s.pdb' % pdb_filename):
run_fetch_pdb(pdb_filename)
pdb_filename = '%s.pdb' % pdb_filename
if 0: # moved adding hydrogens to complete_pdb_hierarchy
if pdb_filename.endswith('.updated.pdb'):
pdb_filename_h = pdb_filename
else:
pdb_filename_h = pdb_filename.replace('.pdb', '.updated.pdb')
if not os.path.exists(pdb_filename_h):
pdb_filename_h = os.path.basename(pdb_filename_h)
print 'pdb_filename_h',pdb_filename_h
if not os.path.exists(pdb_filename_h):
pdb_filename = run_ready_set(pdb_filename)
else: pdb_filename = pdb_filename_h
if 1: # read file option
ppf = hierarchy_utils.get_processed_pdb(pdb_filename=pdb_filename)
elif 0: # raw records example
pdb_inp = iotbx.pdb.input(pdb_filename)
hierarchy = pdb_inp.construct_hierarchy()
raw_records = []
for atom in hierarchy.atoms():
raw_records.append(atom.format_atom_record())
ppf = hierarchy_utils.get_processed_pdb(raw_records=raw_records)
else: # from hierarchy
pdb_inp = iotbx.pdb.input(pdb_filename)
hierarchy = pdb_inp.construct_hierarchy()
ppf = hierarchy_utils.get_processed_pdb(pdb_inp=hierarchy.as_pdb_input())
if not skip_validation:
initial_model_statistics = mmtbx.model.statistics.geometry(
pdb_hierarchy = ppf.all_chain_proxies.pdb_hierarchy)
# should use cctbx
if keep_alt_loc: pass
else:
hierarchy = remove_alternative_locations(
ppf.all_chain_proxies.pdb_hierarchy
)
ppf.all_chain_proxies.pdb_hierarchy.shift_to_origin(
ppf.all_chain_proxies.pdb_inp.crystal_symmetry())
ppf.all_chain_proxies.pdb_hierarchy.remove_residue_groups_with_atoms_on_special_positions_selective(ppf.all_chain_proxies.pdb_inp.crystal_symmetry())
if 0:
output = hierarchy_utils.write_hierarchy(
pdb_filename, # uses to get output filename
ppf.all_chain_proxies.pdb_inp,
hierarchy,
"temp0")
ppf = hierarchy_utils.get_processed_pdb(pdb_filename=output)
else:
raw_records = hierarchy_utils.get_raw_records(
ppf.all_chain_proxies.pdb_inp,
ppf.all_chain_proxies.pdb_hierarchy,
)
ppf = hierarchy_utils.get_processed_pdb(raw_records=raw_records)
hetero_charges = charges.get_hetero_charges(
ppf.all_chain_proxies.pdb_inp,
ppf.all_chain_proxies.pdb_hierarchy,
)
if not hetero_charges:
# some defaults
hetero_charges = charges.default_ion_charges
inter_residue_bonds = charges.get_inter_residue_bonds(ppf)
#
# extends side chains and add hydrogens
#
if model_completion:
use_capping_hydrogens=False
fname = 'complete'
else:
use_capping_hydrogens=True
fname = 'capping'
#assert 0 # model has H
ppf = completion.complete_pdb_hierarchy(
ppf.all_chain_proxies.pdb_hierarchy,
ppf.geometry_restraints_manager(),
pdb_filename=pdb_filename,
pdb_inp=ppf.all_chain_proxies.pdb_inp,
use_capping_hydrogens=use_capping_hydrogens,
)
if calculate_charge:
new_pdb_filename = hierarchy_utils.write_hierarchy(
pdb_filename, # uses to get output filename
ppf.all_chain_proxies.pdb_inp,
ppf.all_chain_proxies.pdb_hierarchy,
fname)
## need now inter_residue_bonds because of added hydrogens
## maybe update the bond table!!!
ppf = hierarchy_utils.get_processed_pdb(pdb_filename=new_pdb_filename)
inter_residue_bonds = charges.get_inter_residue_bonds(ppf, verbose=True)
total_charge = charges.calculate_pdb_hierarchy_charge(
ppf.all_chain_proxies.pdb_hierarchy,
hetero_charges=hetero_charges,
inter_residue_bonds=inter_residue_bonds,
)
print "total_charge",total_charge
## after no error getting total charge, write the completed pdb file
hierarchy_utils.write_hierarchy(pdb_filename, # uses to get output filename
ppf.all_chain_proxies.pdb_inp,
ppf.all_chain_proxies.pdb_hierarchy,
fname)
if not skip_validation:
final_model_statistics = mmtbx.model.statistics.geometry(
pdb_hierarchy = ppf.all_chain_proxies.pdb_hierarchy)
def use_electrons_to_add_hdyrogens(
hierarchy,
geometry_restraints_manager,
use_capping_hydrogens=False,
append_to_end_of_model=False,
):
if not use_capping_hydrogens: return
from qrefine.utils import electrons
rc = []
raw_records = hierarchy_utils.get_raw_records(
pdb_hierarchy=hierarchy,
crystal_symmetry=geometry_restraints_manager.crystal_symmetry,
)
charges = electrons.run(
pdb_filename=None,
raw_records=raw_records,
return_formal_charges=True,
)
charged_atoms = charges.get_charged_atoms()
remove = []
proton_element, proton_name = get_proton_info(hierarchy)
for atom, electrons in charged_atoms:
atom_group = atom.parent()
#if atom_group.resname=='CYS' and atom.name==' SG ':
# if electrons==-1 and atom_group.get_atom('HG'):
# remove.append(atom_group.get_atom('HG'))
if atom.element_is_hydrogen() and electrons == 1:
#print 'REMOVING', atom.quote()
remove.append(atom)
if get_class(atom.parent().resname) in [
'common_amino_acid',
]:
continue
atom = hierarchy.atoms()[atom.i_seq]
# this does not even work
rc = _add_hydrogens_to_atom_group_using_bad(
atom.parent(),
' %s1 ' % proton_element,
proton_element,
atom.name.strip(),
'C4',
'C3',
1.,
120.,
160.,
append_to_end_of_model=append_to_end_of_model,
)
def _atom_i_seq(a1, a2):
if a1.i_seq < a2.i_seq: return -1
return 1
if remove:
remove.sort(key=cmp_to_key(_atom_i_seq))
remove.reverse()
for atom in remove:
# this is a kludge
# print(atom,atom.i_seq)
name = atom.name
atom = hierarchy.atoms()[atom.i_seq]
atom_group = atom.parent()
atom = atom_group.get_atom(name.strip())
atom_group.remove_atom(atom)
return rc
def complete_pdb_hierarchy(
hierarchy,
geometry_restraints_manager,
use_capping_hydrogens=False,
append_to_end_of_model=False,
pdb_filename=None,
pdb_inp=None,
original_pdb_filename=None,
verbose=False,
debug=False,
):
"""Complete PDB hierarchy with hydrogen atoms as needed
This needs to move to Phenix with better functionality
Args:
hierarchy (hierarchy): Starting model
geometry_restraints_manager (GRM): Starting restraints
use_capping_hydrogens (bool, optional): Capping or not
append_to_end_of_model (bool, optional): Added atoms go to end of atom list
pdb_filename (None, optional): Description
pdb_inp (None, optional): Description
original_pdb_filename (None, optional): Description
verbose (bool, optional): Description
debug (bool, optional): Description
Returns:
TYPE: Description
Raises:
Sorry: Description
"""
for ag in hierarchy.atom_groups():
if get_class(ag.resname) in ['common_rna_dna']:
raise Sorry('Nucleotides are not currently supported. e.g. %s' %
ag.resname)
from mmtbx.building import extend_sidechains
original_hierarchy = None
params = hierarchy_utils.get_pdb_interpretation_params()
params.restraints_library.cdl = False
if use_capping_hydrogens:
params.link_distance_cutoff = 1.8 # avoid linking across a single missing AA
if original_pdb_filename:
original_pdb_inp = iotbx.pdb.input(original_pdb_filename)
original_hierarchy = original_pdb_inp.construct_hierarchy()
if debug:
output = hierarchy_utils.write_hierarchy(
pdb_filename,
pdb_inp,
hierarchy,
'temp1',
)
#
# assume model is heavy-atom complete
#
if not use_capping_hydrogens:
if debug:
ppf = hierarchy_utils.get_processed_pdb(pdb_filename=output)
else:
raw_records = hierarchy_utils.get_raw_records(pdb_inp, hierarchy)
ppf = hierarchy_utils.get_processed_pdb(
raw_records=raw_records,
params=params,
)
sites_cart = hierarchy.atoms().extract_xyz()
ppf.all_chain_proxies.pdb_hierarchy.atoms().set_xyz(sites_cart)
n_changed = extend_sidechains.extend_protein_model(
ppf.all_chain_proxies.pdb_hierarchy,
mon_lib_server,
add_hydrogens=False,
)
if debug:
print('number of side chains changed', n_changed)
output = hierarchy_utils.write_hierarchy(
pdb_filename,
pdb_inp,
ppf.all_chain_proxies.pdb_hierarchy,
'temp2',
)
#
# need to use Reduce/ReadySet! to add hydrogens
#
if not use_capping_hydrogens:
output = hierarchy_utils.write_hierarchy(
pdb_filename,
pdb_inp,
ppf.all_chain_proxies.pdb_hierarchy,
'readyset_input',
)
hierarchy = hierarchy_utils.add_hydrogens_using_ReadySet(output)
#
# remove side chain acid hydrogens - maybe not required since recent changes
#
if debug:
ppf = hierarchy_utils.get_processed_pdb(
pdb_filename=output,
params=params,
)
else:
raw_records = hierarchy_utils.get_raw_records(pdb_inp, hierarchy)
ppf = hierarchy_utils.get_processed_pdb(
raw_records=raw_records,
params=params,
)
sites_cart = hierarchy.atoms().extract_xyz()
ppf.all_chain_proxies.pdb_hierarchy.atoms().set_xyz(sites_cart)
remove_acid_side_chain_hydrogens(ppf.all_chain_proxies.pdb_hierarchy)
#
# add hydrogens in special cases
# eg ETA
# eg N - H, H2
#
if debug:
ppf = hierarchy_utils.get_processed_pdb(
pdb_filename=output,
params=params,
)
else:
hierarchy = ppf.all_chain_proxies.pdb_hierarchy
raw_records = hierarchy_utils.get_raw_records(pdb_inp, hierarchy)
ppf = hierarchy_utils.get_processed_pdb(
raw_records=raw_records,
params=params,
)
sites_cart = hierarchy.atoms().extract_xyz()
ppf.all_chain_proxies.pdb_hierarchy.atoms().set_xyz(sites_cart)
special_case_hydrogens(
ppf.all_chain_proxies.pdb_hierarchy,
ppf.geometry_restraints_manager(),
#use_capping_hydrogens=use_capping_hydrogens,
#append_to_end_of_model=append_to_end_of_model,
#original_hierarchy=original_hierarchy,
verbose=verbose,
)
#
# add terminals atoms including hydrogens and OXT - more docs here...
#
if debug:
output = hierarchy_utils.write_hierarchy(
pdb_filename,
pdb_inp,
ppf.all_chain_proxies.pdb_hierarchy,
'temp3',
)
ppf = hierarchy_utils.get_processed_pdb(
pdb_filename=output,
params=params,
)
else:
hierarchy = ppf.all_chain_proxies.pdb_hierarchy
raw_records = hierarchy_utils.get_raw_records(pdb_inp, hierarchy)
ppf = hierarchy_utils.get_processed_pdb(
raw_records=raw_records,
params=params,
)
sites_cart = hierarchy.atoms().extract_xyz()
ppf.all_chain_proxies.pdb_hierarchy.atoms().set_xyz(sites_cart)
#
# moved to mmtbx.ligands
#
add_terminal_hydrogens(
ppf.all_chain_proxies.pdb_hierarchy,
ppf.geometry_restraints_manager(),
use_capping_hydrogens=use_capping_hydrogens,
append_to_end_of_model=append_to_end_of_model,
original_hierarchy=original_hierarchy,
verbose=verbose,
) # in place
ppf.all_chain_proxies.pdb_hierarchy.atoms(
).set_chemical_element_simple_if_necessary()
ppf.all_chain_proxies.pdb_hierarchy.sort_atoms_in_place()
#display_hierarchy_atoms(ppf.all_chain_proxies.pdb_hierarchy)
#ppf.all_chain_proxies.pdb_hierarchy.atoms_reset_serial()
#ppf.all_chain_proxies.pdb_hierarchy.atoms().reset_i_seq()
return ppf
def run(
pdb_filename=None,
pdb_hierarchy=None,
crystal_symmetry=None,
model_completion=True,
original_pdb_filename=None,
append_to_end_of_model=True,
):
#
# function as be used in two main modes
# 1. completing a model with hydrogens in a protein-like manner
# 2. completing a cluster with hydrogens in a QM-sensible manner
#
# Validation
#
if pdb_hierarchy:
assert crystal_symmetry
assert pdb_filename is None
#
# output
#
if model_completion:
use_capping_hydrogens = False
fname = 'complete'
else:
use_capping_hydrogens = True
fname = 'capping'
#
# adjust parameters
#
params = None
if use_capping_hydrogens:
params = hierarchy_utils.get_pdb_interpretation_params()
params.link_distance_cutoff = 1.8
if pdb_hierarchy:
raw_records = hierarchy_utils.get_raw_records(
pdb_inp=None,
pdb_hierarchy=pdb_hierarchy,
crystal_symmetry=crystal_symmetry,
)
ppf = hierarchy_utils.get_processed_pdb(raw_records=raw_records)
sites_cart = pdb_hierarchy.atoms().extract_xyz()
ppf.all_chain_proxies.pdb_hierarchy.atoms().set_xyz(sites_cart)
else:
ppf = hierarchy_utils.get_processed_pdb(
pdb_filename=pdb_filename,
params=params,
)
#
# guts
#
ppf = complete_pdb_hierarchy(
ppf.all_chain_proxies.pdb_hierarchy,
ppf.geometry_restraints_manager(),
use_capping_hydrogens=use_capping_hydrogens,
append_to_end_of_model=append_to_end_of_model, # needed for clustering
# code and Molprobity
pdb_filename=pdb_filename, # used just for naming of debug output
pdb_inp=ppf.all_chain_proxies.pdb_inp, # used in get_raw_records. why
verbose=False,
)
if pdb_filename:
output = hierarchy_utils.write_hierarchy(
pdb_filename,
ppf.all_chain_proxies.pdb_inp,
ppf.all_chain_proxies.pdb_hierarchy,
fname,
)
return ppf.all_chain_proxies.pdb_hierarchy