def generate_ca(self):
t0 = time.time()
self.hierarchy.reset_atom_i_seqs()
self.hierarchy.reset_i_seq_if_necessary()
for five in generate_protein_fragments(
self.hierarchy, self.geometry_restraints_manager, length=5):
rc = []
for atom in five.atoms():
if atom.name == ' CA ':
rc.append(atom)
if len(rc) == 5:
yield rc
def generate_ca(self, length=5):
"""
Geting all C-alpha atoms from an whole pdb file.
"""
# faster with atom selection to CA re CJW update
protain_fragments = generate_protein_fragments(
hierarchy = self.chain_hierarchy,
geometry = self.geometry_restraints_manager,
include_non_standard_peptides=True,
length=length)
for five in protain_fragments:
rc = []
for residue in five:
for atom in residue.atoms():
if atom.name == ' CA ':
rc.append(atom)
if len(rc) == length:
yield rc
def count_residue_sets(pdb_hierarchy=None, length=3, allow_poly_ca=False):
count = 0
resnames = []
for residue_set in generate_protein_fragments(
hierarchy=pdb_hierarchy,
geometry=None,
length=length,
include_non_linked=False,
backbone_only=True,
include_non_standard_peptides=True,
allow_poly_ca=allow_poly_ca,
verbose=False):
count += 1
#resnames = []
#for res in residue_set:
# resnames.append(res.resname+res.resseq)
#resstring = '-'.join(resnames)
#print(resstring)
#print(count)
return count
if __name__ == "__main__":
import sys
from iotbx import pdb
from tst_rdl import get_geometry_restraints_manager
filename = sys.argv[1]
pdb_inp = pdb.input(filename)
pdb_hierarchy = pdb_inp.construct_hierarchy()
geometry_restraints_manager = get_geometry_restraints_manager(filename)
pdb_hierarchy.reset_i_seq_if_necessary()
from mmtbx.conformation_dependent_library import generate_protein_fragments
for i in range(2, 6):
for threes in generate_protein_fragments(
pdb_hierarchy,
geometry_restraints_manager,
length=i,
#verbose=verbose,
):
print threes
try:
print ' omega %5.1f' % threes.get_omega_value()
except:
print ' omega is not valid' # intentional
print ' omegas %s' % threes.get_omega_values()
try:
print " cis? %-5s %s" % (threes.cis_group(),
threes.cis_group(limit=30))
except:
print ' cis? is not valid' # intentional
try:
print " trans? %-5s %s" % (threes.trans_group(),
def __init__(self,
pdb_hierarchy,
nontrans_only=False,
out=sys.stdout,
quiet=True):
validation.__init__(self)
self.residue_count = [0, 0]
#[OMEGA_GENERAL, OMEGA_PRO]
self.omega_count = [[0, 0, 0], [0, 0, 0]]
#[OMEGA_GENERAL, OMEGA_PRO], then
#[OMEGALYZE_TRANS, OMEGALYZE_CIS, OMEGALYZE_TWISTED]
from mmtbx.validation import utils
from scitbx.array_family import flex
self._outlier_i_seqs = flex.size_t()
pdb_atoms = pdb_hierarchy.atoms()
all_i_seqs = pdb_atoms.extract_i_seq()
if all_i_seqs.all_eq(0):
pdb_atoms.reset_i_seq()
use_segids = utils.use_segids_in_place_of_chainids(
hierarchy=pdb_hierarchy)
first_conf_altloc = None
prev_chain_id = None
for twores in generate_protein_fragments(
pdb_hierarchy,
length=2,
geometry=None,
include_non_standard_peptides=True):
main_residue = twores[
1] #this is the relevant residue for id-ing cis-Pro
conf_altloc = get_conformer_altloc(twores)
prevres_altloc, mainres_altloc = get_local_omega_altlocs(twores)
twores_altloc = prevres_altloc or mainres_altloc #default '' evals False
chain = main_residue.parent().parent()
if use_segids:
chain_id = utils.get_segid_as_chainid(chain=chain)
else:
chain_id = chain.id
if chain_id != prev_chain_id: #if we've moved to a new chain...
first_conf_altloc = conf_altloc #...reset reference altloc
prev_chain_id = chain_id
if (conf_altloc != first_conf_altloc) and twores_altloc == '':
#skip non-alternate residues unless this is the first time thru a chain
continue
omega_atoms = get_omega_atoms(twores)
#omega_atoms is the list [CA1 C1 N2 CA2], with None for missing atoms
if None in omega_atoms:
continue
omega = get_omega(omega_atoms)
if omega is None: continue
omega_type = find_omega_type(omega)
if omega_type == OMEGALYZE_TRANS:
is_nontrans = False
else:
is_nontrans = True
self.n_outliers += 1
if main_residue.resname == "PRO": res_type = OMEGA_PRO
else: res_type = OMEGA_GENERAL
self.residue_count[res_type] += 1
self.omega_count[res_type][omega_type] += 1
highest_mc_b = get_highest_mc_b(twores[0].atoms(),
twores[1].atoms())
coords = get_center(main_residue)
markup_atoms = []
for omega_atom in omega_atoms:
markup_atoms.append(
kin_atom(omega_atom.parent().id_str(), omega_atom.xyz))
result = omega_result(
model_id=twores[0].parent().parent().parent().id,
chain_id=chain_id,
resseq=main_residue.resseq,
icode=main_residue.icode,
resname=main_residue.resname,
altloc=mainres_altloc,
prev_resseq=twores[0].resseq,
prev_icode=twores[0].icode,
prev_resname=twores[0].resname,
prev_altloc=prevres_altloc,
segid=None,
omega=omega,
omega_type=omega_type,
res_type=res_type,
is_nontrans=is_nontrans,
outlier=is_nontrans,
highest_mc_b=highest_mc_b,
xyz=coords,
markup_atoms=markup_atoms)
if is_nontrans or not nontrans_only: #(not nontrans_only or is_nontrans)
self.results.append(result)
if is_nontrans:
i_seqs = main_residue.atoms().extract_i_seq()
assert (not i_seqs.all_eq(0)
) #This assert copied from ramalyze
self._outlier_i_seqs.extend(i_seqs)
self.results.sort(key=lambda x: x.model_id + ':' + x.id_str())
def main():
pdb_inp = pdb.input(lines=model_1yjp, source_info='model_1yjp')
pdb_hierarchy = pdb_inp.construct_hierarchy()
geometry_restraints_manager = get_geometry_restraints_manager(
raw_records=model_1yjp)
pdb_hierarchy.reset_i_seq_if_necessary()
from mmtbx.conformation_dependent_library import generate_protein_fragments
for k in range(2, 6):
for j, threes in enumerate(
generate_protein_fragments(
pdb_hierarchy,
geometry_restraints_manager,
length=k,
#verbose=verbose,
)):
i = k - 2
print(i, j, k, threes)
rc = None
try:
rc = threes.get_omega_value()
except:
print(' omega is not valid') # intentional
if i > 0: assert rc == None
else:
print(' omega %5.1f' % rc)
assert approx_equal(
rc, answers['omegas'][j],
0.1), '%f != %f' % (rc, answers['omegas'][j])
rc = threes.get_omega_values()
print(' omegas %s' % rc)
assert approx_equal(rc,
answers['omegas'][j:j + i + 1]), '%s != %s' % (
rc, answers['omegas'][j:j + i + 1])
rc = None
try:
rc = threes.cis_group()
except:
pass # intentional
try:
print(" cis? %-5s %s" % (rc, threes.cis_group(limit=30)))
except:
print(' cis? is not valid') # intentional
if i >= 2:
assert (rc == None or rc == False), '%s!=%s' % (rc, None)
else:
assert rc == False
try:
print(" trans? %-5s %s" %
(threes.trans_group(), threes.trans_group(limit=30)))
except:
print(' tran? is not valid') # intentional
print(' cis/trans/twisted? %s' %
' '.join(threes.cis_trans_twisted_list()))
try:
print(" rama %s" % threes.get_ramalyze_key())
except:
print(' rama not specified') # intentional
print(' conf %s' % threes.is_pure_main_conf())
rc = None
try:
rc = threes.get_phi_psi_angles()
except:
print(' phi/psi not specified') # intentional
print(' phi/psi %s' % rc)
if i < 1: assert rc == None
else:
test = answers['phi_psi'][j * 2:(j + i) * 2]
assert approx_equal(rc[0], test[0]), '%s!=%s' % (rc, test)
assert approx_equal(rc[1], test[1]), '%s!=%s' % (rc, test)
rc = None
try:
rc = threes.get_ca_dihedrals()
except:
print(' CA dihedrals not specified') # intentional
print(' CA dihedrals %s' % rc)
if i <= 1: assert rc == None
else:
test = answers['calphas'][j:j + i + 1 - 2]
assert approx_equal(rc[0], test[0]), '%s != %s' % (rc, test)
print("OK", i + 2)