def get_total_charge(
self,
list_charges=False,
assert_correct_chain_terminii=True,
check=False,
verbose=False,
):
if self.electrons:
from qrefine.utils import electrons
if self.pdb_filename is None and self.raw_records is None:
self.raw_records = hierarchy_utils.get_raw_records(
pdb_hierarchy=self.pdb_hierarchy,
crystal_symmetry=self.crystal_symmetry,
)
charge = electrons.run(
pdb_filename=self.pdb_filename,
raw_records=self.raw_records,
return_formal_charges=list_charges,
verbose=verbose,
)
return charge
assert 0
total_charge = self.calculate_pdb_hierarchy_charge(
self.pdb_hierarchy,
hetero_charges=self.hetero_charges,
inter_residue_bonds=self.inter_residue_bonds,
list_charges=list_charges,
assert_correct_chain_terminii=assert_correct_chain_terminii,
check=check,
verbose=verbose,
)
return total_charge
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
#
if pdb_hierarchy:
assert crystal_symmetry
assert pdb_filename is None
if model_completion:
use_capping_hydrogens=False
fname = 'complete'
else:
use_capping_hydrogens=True
fname = 'capping'
#assert 0 # model has H
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,
)
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?
original_pdb_filename=original_pdb_filename, # used to define breaks in
# main chain for capping
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
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=[]
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(),
' H1 ',
'H',
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(_atom_i_seq)
remove.reverse()
for atom in remove:
# this is a kludge
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 calculate_residue_charge(
self,
rg,
assert_contains_hydrogens=True,
assert_no_alt_loc=True,
hetero_charges=None,
inter_residue_bonds=None,
verbose=False,
):
if self.verbose: verbose = True
if verbose: print '-' * 80
def _terminal(names, check):
for name in check:
if name not in names:
break
else:
return True
return False
def n_terminal(residue_name, atom_names):
if residue_name in ["PRO"]:
check_names = [
[' H2 ', ' H3 '],
[' H 1', ' H 2'], # CHARMM...
[' HN1', ' HN2'], # BABEL...
]
else:
check_names = [
[' H1 ', ' H2 ', ' H3 '],
]
for check_name in check_names:
rc = _terminal(atom_names, check_name)
if rc: break
return rc
def n_capping(residue_name, atom_names):
if residue_name in ["PRO"]:
check_names = [[' H2 ']]
else:
check_names = [
[' H1 ', ' H2 '],
[' H ', ' H2 '], # from finalise
]
for check_name in check_names:
rc = _terminal(atom_names, check_name)
if rc: break
return rc
def nh2_terminal(atom_names):
return _terminal(atom_names, [' HT1', ' HT2'])
def nh3_terminal(atom_names):
return _terminal(atom_names, [' HT1', ' HT2', ' HT3'])
def c_terminal(atom_names):
rc = _terminal(atom_names, [' OXT'])
if not rc: rc = _terminal(atom_names, [' OT1', ' OT2'])
return rc
def c_capping(atom_names):
rc = _terminal(atom_names, [' HC '])
return rc
def covalent_bond(i_seqs, inter_residue_bonds):
for i_seq in i_seqs:
if i_seq in inter_residue_bonds:
return True
return False
############
max_charge = 1
if assert_no_alt_loc:
if len(rg.atom_groups()) > 1:
raise Sorry("alt locs in %s" %
hierarchy_utils.display_residue_group(rg))
# ions
# needs to be centralised!!!
resname = rg.atom_groups()[0].resname
if get_class(resname) == "common_element":
atom = rg.atoms()[0]
if not atom.charge.strip():
if hetero_charges:
charge = hetero_charges.get(atom.parent().resname.strip(),
None)
if charge is None:
raise Sorry(
'no charge found in the model file or hetero_charges for "%s"'
% atom.quote())
else:
return charge, charge
else:
raise Sorry('no charge found in the model file for "%s"' %
atom.quote())
else:
return atom.charge_as_int(), atom.charge_as_int()
# others
hs = 0
atom_names = []
atom_i_seqs = []
for atom in rg.atoms():
if verbose: print '...', atom.quote()
if atom.element_is_hydrogen(): hs += 1
atom_names.append(atom.name)
atom_i_seqs.append(atom.i_seq)
if verbose: print get_class(resname)
if assert_contains_hydrogens:
if hs == 0:
hydrogens = get_aa_polymer_hydrogens(resname)
if len(hydrogens) != 0:
if verbose:
for atom in rg.atoms():
print 'H', atom.quote()
raise Sorry("no hydrogens: %s" %
hierarchy_utils.display_residue_group(rg))
ag = rg.atom_groups()[0]
charge = get_aa_charge(ag.resname)
rc = get_aa_charge(ag.resname)
if ag.resname in ['GLU', 'ASP']:
rc = -1 # reporting only
annot = ''
if verbose:
print '%s\nstarting charge: %s' % ('*' * 80, charge)
if (get_class(
ag.resname) in ["common_amino_acid", "modified_amino_acid"]
or ag.resname in aac.three_letter_l_given_three_letter_d):
if verbose:
print ag.id_str()
print 'number of hydrogens', len(
get_aa_polymer_hydrogens(ag.resname))
poly_hs = len(get_aa_polymer_hydrogens(ag.resname)) - 2
diff_hs = hs - poly_hs
if verbose:
print 'charge: %s poly_hs: %s diff_hs: %s' % (
charge,
poly_hs,
diff_hs,
)
if verbose: print atom_names
if n_terminal(ag.resname, atom_names):
diff_hs -= 1
max_charge += 1
if verbose:
print 'n_terminal'
print 'charge: %s poly_hs: %s diff_hs: %s' % (
charge,
poly_hs,
diff_hs,
)
annot += 'N-term. '
elif nh3_terminal(atom_names):
diff_hs -= 1
max_charge += 1
if verbose: print 'nh3_terminal True'
annot += 'NH3-term. '
elif nh2_terminal(atom_names):
diff_hs -= 1
max_charge += 1
if verbose: print 'nh2_terminal True'
annot += 'NH2-term. '
elif n_capping(ag.resname, atom_names):
diff_hs -= 1
if verbose: print 'n_capping True'
annot += 'N-capp. '
else:
if verbose: print 'no N term'
if c_terminal(atom_names):
diff_hs -= 1
max_charge += 1
if verbose:
print 'c_terminal'
print 'charge: %s poly_hs: %s diff_hs: %s' % (
charge,
poly_hs,
diff_hs,
)
annot += 'C-term. '
elif c_capping(atom_names):
diff_hs -= 1
#max_charge+=1
if verbose:
print 'c_capping'
print 'charge: %s poly_hs: %s diff_hs: %s' % (
charge,
poly_hs,
diff_hs,
)
annot += 'C-capp. '
else:
if verbose: print 'no C term'
if covalent_bond(atom_i_seqs, inter_residue_bonds):
diff_hs += 1
if verbose:
print 'covalent_bond', atom_i_seqs #, inter_residue_bonds
annot += 'Coval. '
if hierarchy_utils.is_n_terminal_atom_group(ag):
if verbose: print 'subtracting due to N terminal'
diff_hs -= 1
if verbose:
print 'residue: %s charge: %s poly_hs: %2s diff_hs: %2s total: %2s %s' % (
ag.resname,
charge,
poly_hs,
diff_hs,
charge + diff_hs,
annot,
)
charge += diff_hs
if charge: verbose = 0
if verbose:
print ' %s charge: %-2s poly_hs: %s diff_hs: %-2s' % (
ag.id_str(),
charge,
poly_hs,
diff_hs,
)
assert abs(
charge
) <= max_charge, 'residue %s charge %s is greater than %s' % (
rg.atoms()[0].quote(),
charge,
max_charge,
)
if resname in allowable_amino_acid_charges:
assert allowable_amino_acid_charges[
resname] - 1 <= charge <= allowable_amino_acid_charges[
resname] + 1, 'resname %s charge %s range %s %s' % (
resname,
charge,
allowable_amino_acid_charges[resname] - 1,
allowable_amino_acid_charges[resname] + 1,
)
else:
restraints = _get_restraints_from_resname(
ag.resname,
self.mon_lib_server,
)
if ag.resname in ['MTN']:
from qrefine.utils import electrons
if self.pdb_filename is None and self.raw_records is None:
self.raw_records = hierarchy_utils.get_raw_records(
pdb_hierarchy=self.pdb_hierarchy,
crystal_symmetry=self.crystal_symmetry,
)
charge = electrons.run(
pdb_filename=self.pdb_filename,
raw_records=self.raw_records,
)
annot = 'non-polymer'
else:
ag_names = set()
for atom in ag.atoms():
ag_names.add(atom.name.strip())
atom_dict = restraints.atom_dict()
cif_names = set()
total = 0
for name, atom in atom_dict.items():
total += atom.partial_charge
cif_names.add(name)
#assert len(cif_names)==len(cif_names.intersection(ag_names))
#assert len(ag_names)==len(cif_names.intersection(ag_names))
assert abs(total-int(total))<0.01, \
'sum of parial charges fo %s not accurate %f' % (ag.name, total)
charge = int(total)
annot = 'ligand'
return charge, rc, annot
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,
):
for ag in hierarchy.atom_groups():
if get_class(ag.resname) in ['common_rna_dna']:
raise Sorry('')
from mmtbx.building import extend_sidechains
params = None
original_hierarchy = None
if use_capping_hydrogens:
params = hierarchy_utils.get_pdb_interpretation_params()
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)
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
