def exercise_2(eps = 1.e-6):
###> Get started from PDB
mon_lib_srv = monomer_library.server.server()
ener_lib = monomer_library.server.ener_lib()
pdb_file = libtbx.env.find_in_repositories(
relative_path="phenix_regression/pdb/phe_abc_tlsanl_out_geometry_minimized.pdb",
test=os.path.isfile)
model = mmtbx.model.manager(
model_input=iotbx.pdb.input(file_name=pdb_file),
build_grm=True)
model.setup_scattering_dictionaries(scattering_table="wk1995")
model.get_xray_structure().convert_to_isotropic()
u_iso_start = model.get_xray_structure().extract_u_iso_or_u_equiv()
model.get_xray_structure().convert_to_anisotropic()
selections = []
selection_strings = ["chain A", "chain B", "chain C"]
for string in selection_strings:
selections.append(model.selection(string = string))
################
selection = flex.bool(model.get_number_of_atoms(), True)
class refinement_flags: pass
refinement_flags.adp_tls = selections
model.set_refinement_flags(refinement_flags)
model.determine_tls_groups(selection_strings=selections, generate_tlsos=selections)
model.set_refinement_flags(refinement_flags)
xray_structure = model.get_xray_structure()
################
###> Get TLS <-> Ucart
T_initial = []
L_initial = []
S_initial = []
T_initial.append([0.11,0.22,0.33,0.12,0.13,0.23])
L_initial.append([1.11,1.22,1.33,1.12,1.13,1.23])
S_initial.append([0.11,0.12,0.13,0.21,0.22,0.23,0.31,0.32,-0.33])
T_initial.append([0.22,0.44,0.66,0.24,0.26,0.46])
L_initial.append([2.22,2.44,2.66,2.24,2.26,2.46])
S_initial.append([0.22,0.24,0.26,0.42,0.44,0.46,0.62,0.64,-0.66])
T_initial.append([0.33,0.66,0.99,0.36,0.39,0.69])
L_initial.append([2.33,2.66,2.99,2.36,2.39,2.69])
S_initial.append([0.22,0.24,0.26,0.42,0.44,0.46,0.62,0.64,-0.66])
tlsosA = tools.generate_tlsos(selections = selections,
xray_structure = xray_structure,
T = T_initial,
L = L_initial,
S = S_initial)
tlsos = tools.generate_tlsos(selections = selections,
xray_structure = xray_structure,
T = T_initial,
L = L_initial,
S = S_initial)
tlsos = tools.make_tlso_compatible_with_u_positive_definite(
tlsos = tlsos,
xray_structure = xray_structure.deep_copy_scatterers(),
selections = selections,
max_iterations = 50,
number_of_u_nonpositive_definite = 0,
eps = eps,
number_of_macro_cycles_for_tls_from_uanisos = 30)
u_cart_answer = tools.u_cart_from_tls(sites_cart = xray_structure.sites_cart(),
selections = selections,
tlsos = tlsos)
xray_structure.scatterers().set_u_cart(xray_structure.unit_cell(),
u_cart_answer)
assert approx_equal(u_cart_answer,
xray_structure.scatterers().extract_u_cart(xray_structure.unit_cell()))
tools.show_tls(tlsos = tlsos, text = "ANSWER")
###> Set up fmodel
sfg_params = mmtbx.f_model.sf_and_grads_accuracy_master_params.extract()
sfg_params.algorithm = "direct"
sfg_params.cos_sin_table = False
dummy = xray_structure.structure_factors(algorithm = sfg_params.algorithm,
d_min = 2.0).f_calc()
f_obs = abs(dummy.structure_factors_from_scatterers(
xray_structure = xray_structure,
algorithm = sfg_params.algorithm,
cos_sin_table = sfg_params.cos_sin_table).f_calc())
flags = f_obs.generate_r_free_flags(fraction=0.01, max_free=2000)
fmodel = mmtbx.f_model.manager(xray_structure = xray_structure,
f_obs = f_obs,
r_free_flags = flags,
target_name = "ls_wunit_k1",
sf_and_grads_accuracy_params = sfg_params)
fmodel.info(free_reflections_per_bin=250, max_number_of_bins=30).show_all()
xray_structure.convert_to_isotropic()
xray_structure.set_b_iso(value = 25.0)
fmodel.update_xray_structure(xray_structure = xray_structure,
update_f_calc = True)
fmodel.info(free_reflections_per_bin=250, max_number_of_bins=30).show_all()
print("*"*80)
###> TLS refinement against xray data
if (not "--comprehensive" in sys.argv[1:]):
number_of_macro_cycles = 1
max_number_of_iterations = 3
else:
number_of_macro_cycles = 100
max_number_of_iterations = 50
for start_tls_value in [None]:#[0.0, tlsosA, None]:
#for start_tls_value in [None]:
print(" \n "+str(start_tls_value) + " \n ")
fmodel_cp = fmodel.deep_copy()
#for sc in fmodel_cp.xray_structure.scatterers():
# sc.flags.set_use_u_aniso(True)
fmodel_cp.xray_structure.convert_to_anisotropic()
if(start_tls_value is None):
run_finite_differences_test = True
else: run_finite_differences_test = False
model.set_xray_structure(fmodel_cp.xray_structure)
tls_refinement_manager = tools.tls_refinement(
fmodel = fmodel_cp,
model = model,
selections = selections,
selections_1d = None,
refine_T = 1,
refine_L = 1,
refine_S = 1,
number_of_macro_cycles = number_of_macro_cycles,
max_number_of_iterations = max_number_of_iterations,
start_tls_value = start_tls_value,
run_finite_differences_test = run_finite_differences_test,
eps = eps)
u_cart = tls_refinement_manager.fmodel.xray_structure.scatterers().extract_u_cart(
xray_structure.unit_cell())
if("--comprehensive" in sys.argv[1:]):
format = "%10.6f %10.6f %10.6f %10.6f %10.6f %10.6f"
counter = 0
if(start_tls_value == tlsosA): tolerance = 1.e-6
else: tolerance = 0.02
for m1,m2 in zip(u_cart_answer, u_cart):
counter += 1
if(counter < 10):
print("1=" + format % (m1[0],m1[1],m1[2],m1[3],m1[4],m1[5]))
print("2=" + format % (m2[0],m2[1],m2[2],m2[3],m2[4],m2[5]))
assert approx_equal(m1,m2, tolerance)
def uaniso_from_tls_and_back():
mon_lib_srv = monomer_library.server.server()
ener_lib = monomer_library.server.ener_lib()
pdb_file = libtbx.env.find_in_repositories(
relative_path="phenix_regression/pdb/1OC2_tst.pdb",
test=os.path.isfile)
processed_pdb_file = monomer_library.pdb_interpretation.process(
mon_lib_srv = mon_lib_srv,
ener_lib = ener_lib,
file_name = pdb_file,
raw_records = None,
force_symmetry = True)
xray_structure = processed_pdb_file.xray_structure()
selections = []
for string in ["chain A", "chain B"]:
selections.append(processed_pdb_file.all_chain_proxies.selection(
string = string))
input_tls_data = iotbx.pdb.remark_3_interpretation.extract_tls_parameters(
remark_3_records=processed_pdb_file.all_chain_proxies.pdb_inp
.extract_remark_iii_records(iii=3),
pdb_hierarchy = processed_pdb_file.all_chain_proxies.pdb_hierarchy,
chain_ids=[]).tls_params
tls_params = []
for item in input_tls_data:
tls_params.append(tools.tlso(t = item.t,
l = item.l,
s = item.s,
origin = item.origin))
tools.show_tls(tlsos = tls_params)
u_cart_from_tls = tools.u_cart_from_tls(
sites_cart = xray_structure.sites_cart(),
selections = selections,
tlsos = tls_params)
i = 0
for utls,atom in zip(u_cart_from_tls,
processed_pdb_file.all_chain_proxies.pdb_atoms):
updb = atom.uij
#i += 1
#print " ", i
#print "%6.4f %6.4f %6.4f %6.4f %6.4f %6.4f "% \
# (utls[0],utls[1],utls[2],utls[3],utls[4],utls[5])
#print "%6.4f %6.4f %6.4f %6.4f %6.4f %6.4f "% \
# (updb[0],updb[1],updb[2],updb[3],updb[4],updb[5])
assert approx_equal(utls,updb, 1.e-4)
tlsos_initial = []
for input_tls_data_ in input_tls_data:
tlsos_initial.append(tools.tlso(
t = ([0.0,0.0,0.0,0.0,0.0,0.0]),
l = ([0.0,0.0,0.0,0.0,0.0,0.0]),
s = ([0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0]),
origin = input_tls_data_.origin))
tls_from_uanisos = tools.tls_from_uanisos(
number_of_macro_cycles = 300,
max_iterations = 1000,
xray_structure = xray_structure,
selections = selections,
tlsos_initial = tlsos_initial)
print "\nTLS from Uaniso:\n"
tools.show_tls(tlsos = tls_from_uanisos)
for input_tls_data_,tls_from_uanisos_ in zip(tls_params,tls_from_uanisos):
assert approx_equal(input_tls_data_.t, tls_from_uanisos_.t, 1.e-4)
assert approx_equal(input_tls_data_.l, tls_from_uanisos_.l, 1.e-4)
assert approx_equal(input_tls_data_.s, tls_from_uanisos_.s, 1.e-4)
assert approx_equal(input_tls_data_.origin, tls_from_uanisos_.origin, 1.e-3)
#
print format_cpu_times()
def uaniso_from_tls_and_back():
mon_lib_srv = monomer_library.server.server()
ener_lib = monomer_library.server.ener_lib()
pdb_file = libtbx.env.find_in_repositories(
relative_path="phenix_regression/pdb/1OC2_tst.pdb",
test=os.path.isfile)
processed_pdb_file = monomer_library.pdb_interpretation.process(
mon_lib_srv = mon_lib_srv,
ener_lib = ener_lib,
file_name = pdb_file,
raw_records = None,
force_symmetry = True)
xray_structure = processed_pdb_file.xray_structure()
selections = []
for string in ["chain A", "chain B"]:
selections.append(processed_pdb_file.all_chain_proxies.selection(
string = string))
input_tls_data = iotbx.pdb.remark_3_interpretation.extract_tls_parameters(
remark_3_records=processed_pdb_file.all_chain_proxies.pdb_inp
.extract_remark_iii_records(iii=3),
pdb_hierarchy = processed_pdb_file.all_chain_proxies.pdb_hierarchy,
chain_ids=[]).tls_params
tls_params = []
for item in input_tls_data:
tls_params.append(tools.tlso(t = item.t,
l = item.l,
s = item.s,
origin = item.origin))
tools.show_tls(tlsos = tls_params)
u_cart_from_tls = tools.u_cart_from_tls(
sites_cart = xray_structure.sites_cart(),
selections = selections,
tlsos = tls_params)
i = 0
for utls,atom in zip(u_cart_from_tls,
processed_pdb_file.all_chain_proxies.pdb_atoms):
updb = atom.uij
#i += 1
#print " ", i
#print "%6.4f %6.4f %6.4f %6.4f %6.4f %6.4f "% \
# (utls[0],utls[1],utls[2],utls[3],utls[4],utls[5])
#print "%6.4f %6.4f %6.4f %6.4f %6.4f %6.4f "% \
# (updb[0],updb[1],updb[2],updb[3],updb[4],updb[5])
assert approx_equal(utls,updb, 1.e-4)
tlsos_initial = []
for input_tls_data_ in input_tls_data:
tlsos_initial.append(tools.tlso(
t = ([0.0,0.0,0.0,0.0,0.0,0.0]),
l = ([0.0,0.0,0.0,0.0,0.0,0.0]),
s = ([0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0]),
origin = input_tls_data_.origin))
tls_from_uanisos = tools.tls_from_uanisos(
number_of_macro_cycles = 300,
max_iterations = 1000,
xray_structure = xray_structure,
selections = selections,
tlsos_initial = tlsos_initial)
print("\nTLS from Uaniso:\n")
tools.show_tls(tlsos = tls_from_uanisos)
for input_tls_data_,tls_from_uanisos_ in zip(tls_params,tls_from_uanisos):
assert approx_equal(input_tls_data_.t, tls_from_uanisos_.t, 1.e-4)
assert approx_equal(input_tls_data_.l, tls_from_uanisos_.l, 1.e-4)
assert approx_equal(input_tls_data_.s, tls_from_uanisos_.s, 1.e-4)
assert approx_equal(input_tls_data_.origin, tls_from_uanisos_.origin, 1.e-3)
#
print(format_cpu_times())
def exercise_2(eps = 1.e-6):
###> Get started from PDB
mon_lib_srv = monomer_library.server.server()
ener_lib = monomer_library.server.ener_lib()
pdb_file = libtbx.env.find_in_repositories(
relative_path="phenix_regression/pdb/phe_abc_tlsanl_out_geometry_minimized.pdb",
test=os.path.isfile)
processed_pdb_file = monomer_library.pdb_interpretation.process(
mon_lib_srv = mon_lib_srv,
ener_lib = ener_lib,
file_name = pdb_file,
raw_records = None,
force_symmetry = True)
xray_structure = processed_pdb_file.xray_structure()
xray_structure.scattering_type_registry(table = "wk1995")
xray_structure.convert_to_isotropic()
u_iso_start = xray_structure.extract_u_iso_or_u_equiv()
xray_structure.convert_to_anisotropic()
selections = []
selection_strings = ["chain A", "chain B", "chain C"]
for string in selection_strings:
selections.append(processed_pdb_file.all_chain_proxies.selection(
string = string))
################
geometry = processed_pdb_file.geometry_restraints_manager(
show_energies = False,
plain_pairs_radius = 5.0)
restraints_manager = mmtbx.restraints.manager(geometry = geometry,
normalization = False)
selection = flex.bool(xray_structure.scatterers().size(), True)
class refinement_flags: pass
refinement_flags.adp_tls = selections
tls_groups = tools.tls_groups(selection_strings = selection_strings)
tlsos = tools.generate_tlsos(
selections = refinement_flags.adp_tls,
xray_structure = xray_structure,
value = 0.0)
tls_groups.tlsos = tlsos
model = mmtbx.model.manager(
refinement_flags = refinement_flags,
restraints_manager = restraints_manager,
xray_structure = xray_structure,
tls_groups = tls_groups,
pdb_hierarchy = processed_pdb_file.all_chain_proxies.pdb_hierarchy)
################
###> Get TLS <-> Ucart
T_initial = []
L_initial = []
S_initial = []
T_initial.append([0.11,0.22,0.33,0.12,0.13,0.23])
L_initial.append([1.11,1.22,1.33,1.12,1.13,1.23])
S_initial.append([0.11,0.12,0.13,0.21,0.22,0.23,0.31,0.32,-0.33])
T_initial.append([0.22,0.44,0.66,0.24,0.26,0.46])
L_initial.append([2.22,2.44,2.66,2.24,2.26,2.46])
S_initial.append([0.22,0.24,0.26,0.42,0.44,0.46,0.62,0.64,-0.66])
T_initial.append([0.33,0.66,0.99,0.36,0.39,0.69])
L_initial.append([2.33,2.66,2.99,2.36,2.39,2.69])
S_initial.append([0.22,0.24,0.26,0.42,0.44,0.46,0.62,0.64,-0.66])
tlsosA = tools.generate_tlsos(selections = selections,
xray_structure = xray_structure,
T = T_initial,
L = L_initial,
S = S_initial)
tlsos = tools.generate_tlsos(selections = selections,
xray_structure = xray_structure,
T = T_initial,
L = L_initial,
S = S_initial)
tlsos = tools.make_tlso_compatible_with_u_positive_definite(
tlsos = tlsos,
xray_structure = xray_structure.deep_copy_scatterers(),
selections = selections,
max_iterations = 50,
number_of_u_nonpositive_definite = 0,
eps = eps,
number_of_macro_cycles_for_tls_from_uanisos = 30)
u_cart_answer = tools.u_cart_from_tls(sites_cart = xray_structure.sites_cart(),
selections = selections,
tlsos = tlsos)
xray_structure.scatterers().set_u_cart(xray_structure.unit_cell(),
u_cart_answer)
assert approx_equal(u_cart_answer,
xray_structure.scatterers().extract_u_cart(xray_structure.unit_cell()))
tools.show_tls(tlsos = tlsos, text = "ANSWER")
###> Set up fmodel
sfg_params = mmtbx.f_model.sf_and_grads_accuracy_master_params.extract()
sfg_params.algorithm = "direct"
sfg_params.cos_sin_table = False
dummy = xray_structure.structure_factors(algorithm = sfg_params.algorithm,
d_min = 2.0).f_calc()
f_obs = abs(dummy.structure_factors_from_scatterers(
xray_structure = xray_structure,
algorithm = sfg_params.algorithm,
cos_sin_table = sfg_params.cos_sin_table).f_calc())
flags = f_obs.generate_r_free_flags(fraction=0.01, max_free=2000)
fmodel = mmtbx.f_model.manager(xray_structure = xray_structure,
f_obs = f_obs,
r_free_flags = flags,
target_name = "ls_wunit_k1",
sf_and_grads_accuracy_params = sfg_params)
fmodel.info(free_reflections_per_bin=250, max_number_of_bins=30).show_all()
xray_structure.convert_to_isotropic()
xray_structure.set_b_iso(value = 25.0)
fmodel.update_xray_structure(xray_structure = xray_structure,
update_f_calc = True)
fmodel.info(free_reflections_per_bin=250, max_number_of_bins=30).show_all()
print "*"*80
###> TLS refinement against xray data
if (not "--comprehensive" in sys.argv[1:]):
number_of_macro_cycles = 1
max_number_of_iterations = 3
else:
number_of_macro_cycles = 100
max_number_of_iterations = 50
for start_tls_value in [None]:#[0.0, tlsosA, None]:
#for start_tls_value in [None]:
print " \n "+str(start_tls_value) + " \n "
fmodel_cp = fmodel.deep_copy()
#for sc in fmodel_cp.xray_structure.scatterers():
# sc.flags.set_use_u_aniso(True)
fmodel_cp.xray_structure.convert_to_anisotropic()
if(start_tls_value is None):
run_finite_differences_test = True
else: run_finite_differences_test = False
model.xray_structure = fmodel_cp.xray_structure
tls_refinement_manager = tools.tls_refinement(
fmodel = fmodel_cp,
model = model,
selections = selections,
selections_1d = None,
refine_T = 1,
refine_L = 1,
refine_S = 1,
number_of_macro_cycles = number_of_macro_cycles,
max_number_of_iterations = max_number_of_iterations,
start_tls_value = start_tls_value,
run_finite_differences_test = run_finite_differences_test,
eps = eps)
u_cart = tls_refinement_manager.fmodel.xray_structure.scatterers().extract_u_cart(
xray_structure.unit_cell())
if("--comprehensive" in sys.argv[1:]):
format = "%10.6f %10.6f %10.6f %10.6f %10.6f %10.6f"
counter = 0
if(start_tls_value == tlsosA): tolerance = 1.e-6
else: tolerance = 0.02
for m1,m2 in zip(u_cart_answer, u_cart):
counter += 1
if(counter < 10):
print "1=" + format % (m1[0],m1[1],m1[2],m1[3],m1[4],m1[5])
print "2=" + format % (m2[0],m2[1],m2[2],m2[3],m2[4],m2[5])
assert approx_equal(m1,m2, tolerance)
