def export_single_mad_hkl(params):
data = sorted(read_input_data("mad", params),
key=lambda d: d.info().wavelength)
dt_labels = {
v: k
for (k, v) in params.items() if k in ("peak", "infl", "hrem", "lrem")
}
if not dt_labels:
dt_labels = dict(
zip(
("".join(dt.info().labels) for dt in data),
("hrem", "peak", "infl", "lrem"),
))
res = {}
for dt in data:
for dt_name in dt_labels:
if dt_name in "".join(dt.info().labels):
dt_label = dt_labels[dt_name]
hkl_filename = "%s_%s.hkl" % (params["prefix"], dt_label)
print("Exporting data from columns %s to %s" %
(pformat(dt.info().labels), hkl_filename))
with open(hkl_filename, "w") as fp:
miller_array_export_as_shelx_hklf(dt, fp)
res[dt_label] = hkl_filename
return res
def export_sad_hkl(params):
data = read_input_data("sad", params)
if len(data) > 1:
try:
sel_data = next(dt for dt in data
if params["label"] in "".join(dt.info().labels))
except KeyError:
raise ValueError(
"Please specify column label to select a single dataset using --label parameter"
)
except StopIteration:
raise ValueError("Dataset matching column label %s not found" %
params["label"])
else:
sel_data = data[0]
if "label" in params:
sel_data_labels = "".join(sel_data.info().labels)
if params["label"] not in sel_data_labels:
raise ValueError(
"Selected dataset colum label doesn't match --label %s" %
params["label"])
hkl_filename = "%s.hkl" % params["prefix"]
print("Exporting data from columns %s to %s" %
(pformat(sel_data.info().labels), hkl_filename))
with open(hkl_filename, "w") as fp:
miller_array_export_as_shelx_hklf(sel_data, fp)
return {"sad": hkl_filename}
def to_shelxcde(hklin, prefix, sites=0):
'''Read hklin (unmerged reflection file) and generate SHELXC input file
and HKL file'''
from iotbx.reflection_file_reader import any_reflection_file
from iotbx.shelx.hklf import miller_array_export_as_shelx_hklf
reader = any_reflection_file(hklin)
intensities = [
ma for ma in reader.as_miller_arrays(merge_equivalents=False)
if ma.info().labels == ['I', 'SIGI']
][0]
mtz_object = reader.file_content()
indices = reader.file_content().extract_original_index_miller_indices()
intensities = intensities.customized_copy(indices=indices,
info=intensities.info())
with open('%s.hkl' % prefix, 'wb') as hkl_file_handle:
miller_array_export_as_shelx_hklf(intensities, hkl_file_handle)
uc = intensities.unit_cell().parameters()
sg = intensities.space_group().type().lookup_symbol().replace(' ', '')
open('%s.sh' % prefix, 'w').write('\n'.join([
'shelxc %s << eof' % prefix,
'cell %f %f %f %f %f %f' % uc,
'spag %s' % sg,
'sad %s.hkl' % prefix,
'find %d' % sites,
'maxm %d' % ((2 * intensities.data().size() // 1000000) + 1), 'eof', ''
]))
def export_mad_hkl(params):
res = {}
for lb in ("peak", "infl", "hrem", "lrem"):
if lb in params:
data = read_input_data(lb, params)
if len(data) > 1:
raise ValueError("Multiple datasets found in %s" % params[lb])
hkl_filename = "%s_%s.hkl" % (params["prefix"], lb)
with open(hkl_filename, "w") as fp:
miller_array_export_as_shelx_hklf(data[0], fp)
res[lb] = hkl_filename
return res
def export_native_hkl(params):
data = read_input_data("nat", params, False)
if len(data) > 1:
raise ValueError(
"Multiple datasets found in the input native data file %s" %
params["nat"])
hkl_filename = "%s_nat.hkl" % params["prefix"]
print("Exporting data from columns %s to %s" %
(pformat(data[0].info().labels), hkl_filename))
with open(hkl_filename, "w") as fp:
miller_array_export_as_shelx_hklf(data[0], fp)
return {"nat": hkl_filename}
def to_amber(hklin, prefix):
"""Read hklin (unmerged reflection file) HKL file containing
H K L I sqrt(|I|)"""
from iotbx.reflection_file_reader import any_reflection_file
from iotbx.shelx import writer
from iotbx.shelx.hklf import miller_array_export_as_shelx_hklf
reader = any_reflection_file(hklin)
mtz_object = reader.file_content()
ma = reader.as_miller_arrays(merge_equivalents=False)
labels = [c.info().labels for c in ma]
if ["I", "SIGI"] not in labels:
if ["I(+)", "SIGI(+)", "I(-)", "SIGI(-)"] in labels:
print("Error: xia2.to_amber must be run on unmerged data.")
else:
print("Error: columns I / SIGI not found.")
sys.exit(1)
intensities = [c for c in ma if c.info().labels == ["I", "SIGI"]][0]
indices = reader.file_content().extract_original_index_miller_indices()
intensities = intensities.customized_copy(indices=indices,
info=intensities.info())
from scitbx.array_family import flex
abs_intensities = flex.sqrt(flex.abs(intensities.data()))
intensities.set_sigmas(abs_intensities)
with open("%s.hkl" % prefix, "wb") as hkl_file_handle:
miller_array_export_as_shelx_hklf(
intensities,
hkl_file_handle,
scale_range=(-9999.0, 9999.0),
normalise_if_format_overflow=True,
)
return
def to_shelxcde(hklin, prefix, sites=0):
"""Read hklin (unmerged reflection file) and generate SHELXC input file
and HKL file"""
from iotbx.reflection_file_reader import any_reflection_file
from iotbx.shelx.hklf import miller_array_export_as_shelx_hklf
reader = any_reflection_file(hklin)
intensities = [
ma for ma in reader.as_miller_arrays(merge_equivalents=False)
if ma.info().labels == ["I", "SIGI"]
][0]
mtz_object = reader.file_content()
indices = reader.file_content().extract_original_index_miller_indices()
intensities = intensities.customized_copy(indices=indices,
info=intensities.info())
with open("%s.hkl" % prefix, "wb") as hkl_file_handle:
miller_array_export_as_shelx_hklf(intensities, hkl_file_handle)
uc = intensities.unit_cell().parameters()
sg = intensities.space_group().type().lookup_symbol().replace(" ", "")
# spacegroup name disputes
if sg.startswith("R3:"):
sg = "H3"
elif sg.startswith("R32:"):
sg = "H32"
open("%s.sh" % prefix, "w").write("\n".join([
"shelxc %s << eof" % prefix,
"cell %f %f %f %f %f %f" % uc,
"spag %s" % sg,
"sad %s.hkl" % prefix,
"find %d" % sites,
"maxm %d" % ((2 * intensities.data().size() // 1000000) + 1),
"eof",
"",
]))
def to_shelxcde(hklin, prefix, sites=0):
'''Read hklin (unmerged reflection file) and generate SHELXC input file
and HKL file'''
from iotbx.reflection_file_reader import any_reflection_file
from iotbx.shelx.hklf import miller_array_export_as_shelx_hklf
reader = any_reflection_file(hklin)
intensities = [ma for ma in reader.as_miller_arrays(merge_equivalents=False)
if ma.info().labels == ['I', 'SIGI']][0]
mtz_object = reader.file_content()
indices = reader.file_content().extract_original_index_miller_indices()
intensities = intensities.customized_copy(indices=indices, info=intensities.info())
with open('%s.hkl' % prefix, 'wb') as hkl_file_handle:
miller_array_export_as_shelx_hklf(intensities, hkl_file_handle)
uc = intensities.unit_cell().parameters()
sg = intensities.space_group().type().lookup_symbol().replace(' ', '')
open('%s.sh' % prefix, 'w').write('\n'.join([
'shelxc %s << eof' % prefix,
'cell %f %f %f %f %f %f' % uc,
'spag %s' % sg,
'sad %s.hkl' % prefix,
'find %d' % sites,
'maxm %d' % ((2 * intensities.data().size() // 1000000) + 1),
'eof', '']))
def to_shelx(hklin, prefix, compound="", options=None):
"""Read hklin (unmerged reflection file) and generate SHELXT input file
and HKL file"""
reader = any_reflection_file(hklin)
mtz_object = reader.file_content()
ma = reader.as_miller_arrays(merge_equivalents=False)
labels = [c.info().labels for c in ma]
if ["I", "SIGI"] not in labels:
if ["I(+)", "SIGI(+)", "I(-)", "SIGI(-)"] in labels:
print("Error: xia2.to_shelx must be run on unmerged data.")
else:
print("Error: columns I / SIGI not found.")
sys.exit(1)
intensities = [c for c in ma if c.info().labels == ["I", "SIGI"]][0]
indices = reader.file_content().extract_original_index_miller_indices()
intensities = intensities.customized_copy(indices=indices,
info=intensities.info())
with open("%s.hkl" % prefix, "w") as hkl_file_handle:
# limit values to 4 digits (before decimal point), as this is what shelxt
# writes in its output files, and shelxl seems to read. ShelXL apparently
# does not read values >9999 properly
miller_array_export_as_shelx_hklf(
intensities,
hkl_file_handle,
scale_range=(-9999.0, 9999.0),
normalise_if_format_overflow=True,
)
crystal_symm = intensities.crystal_symmetry()
wavelength = None
if options:
wavelength = options.wavelength
if wavelength is None:
mtz_crystals = mtz_object.crystals()
wavelength = mtz_crystals[1].datasets()[0].wavelength()
print("Experimental wavelength: %.3f Angstroms" % wavelength)
unit_cell_dims = None
unit_cell_esds = None
cell_data = None
if options and options.cell:
if options.cell.endswith((".expt", ".json")):
with open(options.cell, "r") as fh:
cell_data = json.load(fh)
if "solution_constrained" in cell_data:
# json from xia2.get_unit_cell_errors (obsolete)
solution = cell_data.get("solution_constrained",
cell_data["solution_unconstrained"])
unit_cell_dims = tuple([
solution[dim]["mean"]
for dim in ["a", "b", "c", "alpha", "beta", "gamma"]
])
unit_cell_esds = tuple([
solution[dim]["population_standard_deviation"]
for dim in ["a", "b", "c", "alpha", "beta", "gamma"]
])
else:
# json from dials.two_theta_refine
cell_data = None
from dxtbx.model.experiment.experiment_list import ExperimentListFactory
experiments = ExperimentListFactory.from_json_file(
options.cell)
crystal = experiments.crystals()[0]
unit_cell_dims = crystal.get_unit_cell().parameters()
unit_cell_esds = crystal.get_cell_parameter_sd()
else: # treat as .cif
import iotbx.cif
cif = iotbx.cif.reader(file_path=options.cell).model()
unit_cell = [
cif.get("xia2", cif.get("two_theta_refine", {})).get(key)
for key in (
"_cell_length_a",
"_cell_length_b",
"_cell_length_c",
"_cell_angle_alpha",
"_cell_angle_beta",
"_cell_angle_gamma",
)
]
# now need to inverse transform these ESD numbers into unit_cell_dims and unit_cell_esds
def dim_esd(iucr_number_format):
if "(" not in iucr_number_format:
return float(iucr_number_format), 0
p = iucr_number_format.split("(")
dim = float(p[0])
esd = float(p[1].split(")")[0])
if "." in p[0]:
esd = esd / (10**len(p[0].split(".")[1]))
return dim, esd
unit_cell_dims, unit_cell_esds = list(
zip(*(dim_esd(p) for p in unit_cell)))
cb_op = crystal_symm.change_of_basis_op_to_reference_setting()
if cb_op.c().r().as_hkl() == "h,k,l":
print("Change of basis to reference setting: %s" % cb_op)
crystal_symm = crystal_symm.change_basis(cb_op)
if str(cb_op) != "a,b,c":
unit_cell_dims = None
unit_cell_esds = None
# Would need to apply operation to cell errors, too. Need a test case for this
# crystal_symm.show_summary()
xray_structure = structure(crystal_symmetry=crystal_symm)
if compound:
result = parse_compound(compound)
for element in result:
xray_structure.add_scatterer(
scatterer(label=element, occupancy=result[element]))
open("%s.ins" % prefix, "w").write("".join(
writer.generator(
xray_structure,
wavelength=wavelength,
full_matrix_least_squares_cycles=0,
title=prefix,
unit_cell_dims=unit_cell_dims,
unit_cell_esds=unit_cell_esds,
)))
if options and options.cell and not options.cell.endswith(".cif"):
generate_cif(
prefix=prefix,
unit_cell_data=cell_data,
wavelength=wavelength,
structure=xray_structure,
)
def to_shelx(hklin, prefix, compound='', options=None):
'''Read hklin (unmerged reflection file) and generate SHELXT input file
and HKL file'''
from iotbx.reflection_file_reader import any_reflection_file
from iotbx.shelx import writer
from iotbx.shelx.hklf import miller_array_export_as_shelx_hklf
from cctbx.xray.structure import structure
from cctbx.xray import scatterer
reader = any_reflection_file(hklin)
mtz_object = reader.file_content()
ma = reader.as_miller_arrays(merge_equivalents=False)
labels = [c.info().labels for c in ma]
if ['I', 'SIGI'] not in labels:
if ['I(+)', 'SIGI(+)', 'I(-)', 'SIGI(-)'] in labels:
print "Error: xia2.to_shelx must be run on unmerged data."
else:
print "Error: columns I / SIGI not found."
sys.exit(1)
intensities = [c for c in ma if c.info().labels == ['I', 'SIGI']][0]
# FIXME do I need to reindex to a conventional setting here
indices = reader.file_content().extract_original_index_miller_indices()
intensities = intensities.customized_copy(indices=indices,
info=intensities.info())
with open('%s.hkl' % prefix, 'wb') as hkl_file_handle:
# limit values to 4 digits (before decimal point), as this is what shelxt
# writes in its output files, and shelxl seems to read. ShelXL apparently
# does not read values >9999 properly
miller_array_export_as_shelx_hklf(intensities,
hkl_file_handle,
scale_range=(-9999., 9999.),
normalise_if_format_overflow=True)
crystal_symm = intensities.crystal_symmetry()
wavelength = None
if options:
wavelength = options.wavelength
if wavelength is None:
mtz_crystals = mtz_object.crystals()
wavelength = mtz_crystals[1].datasets()[0].wavelength()
print 'Experimental wavelength: %.3f Angstroms' % wavelength
unit_cell_dims = None
unit_cell_esds = None
cell_data = None
if options and options.cell:
if options.cell.endswith('.json'):
with open(options.cell, 'r') as fh:
cell_data = json.load(fh)
if 'solution_constrained' in cell_data:
# json from xia2.get_unit_cell_errors (obsolete)
solution = cell_data.get('solution_constrained',
cell_data['solution_unconstrained'])
unit_cell_dims = tuple([
solution[dim]['mean']
for dim in ['a', 'b', 'c', 'alpha', 'beta', 'gamma']
])
unit_cell_esds = tuple([
solution[dim]['population_standard_deviation']
for dim in ['a', 'b', 'c', 'alpha', 'beta', 'gamma']
])
else:
# json from dials.two_theta_refine
cell_data = None
from dxtbx.model.experiment.experiment_list import ExperimentListFactory
experiments = ExperimentListFactory.from_json_file(
options.cell)
crystal = experiments.crystals()[0]
unit_cell_dims = crystal.get_unit_cell().parameters()
unit_cell_esds = crystal.get_cell_parameter_sd()
else: # treat as .cif
import iotbx.cif
cif = iotbx.cif.reader(file_path=options.cell).model()
unit_cell = [
cif.get('xia2', cif.get('two_theta_refine', {})).get(key)
for key in ('_cell_length_a', '_cell_length_b',
'_cell_length_c', '_cell_angle_alpha',
'_cell_angle_beta', '_cell_angle_gamma')
]
# now need to inverse transform these ESD numbers into unit_cell_dims and unit_cell_esds
def dim_esd(iucr_number_format):
if '(' not in iucr_number_format:
return float(iucr_number_format), 0
p = iucr_number_format.split('(')
dim = float(p[0])
esd = float(p[1].split(')')[0])
if '.' in p[0]:
esd = esd / (10**len(p[0].split('.')[1]))
return dim, esd
unit_cell_dims, unit_cell_esds = zip(*(dim_esd(p)
for p in unit_cell))
cb_op = crystal_symm.change_of_basis_op_to_reference_setting()
if cb_op.c().r().as_hkl() == 'h,k,l':
print 'Change of basis to reference setting: %s' % cb_op
crystal_symm = crystal_symm.change_basis(cb_op)
if str(cb_op) != "a,b,c":
unit_cell_dims = None
unit_cell_esds = None
# Would need to apply operation to cell errors, too. Need a test case for this
# crystal_symm.show_summary()
xray_structure = structure(crystal_symmetry=crystal_symm)
if compound:
result = parse_compound(compound)
for element in result:
xray_structure.add_scatterer(
scatterer(label=element, occupancy=result[element]))
open('%s.ins' % prefix, 'w').write(''.join(
writer.generator(xray_structure,
wavelength=wavelength,
full_matrix_least_squares_cycles=0,
title=prefix,
unit_cell_dims=unit_cell_dims,
unit_cell_esds=unit_cell_esds)))
if options and options.cell and not options.cell.endswith('.cif'):
generate_cif(prefix=prefix,
unit_cell_data=cell_data,
wavelength=wavelength,
structure=xray_structure)
def to_shelx(hklin, prefix, compound='', options={}):
'''Read hklin (unmerged reflection file) and generate SHELXT input file
and HKL file'''
from iotbx.reflection_file_reader import any_reflection_file
from iotbx.shelx import writer
from iotbx.shelx.hklf import miller_array_export_as_shelx_hklf
from cctbx.xray.structure import structure
from cctbx.xray import scatterer
reader = any_reflection_file(hklin)
mtz_object = reader.file_content()
intensities = [ma for ma in reader.as_miller_arrays(merge_equivalents=False)
if ma.info().labels == ['I', 'SIGI']][0]
# FIXME do I need to reindex to a conventional setting here
indices = reader.file_content().extract_original_index_miller_indices()
intensities = intensities.customized_copy(indices=indices, info=intensities.info())
with open('%s.hkl' % prefix, 'wb') as hkl_file_handle:
# limit values to 4 digits (before decimal point), as this is what shelxt
# writes in its output files, and shelxl seems to read. ShelXL apparently
# does not read values >9999 properly
miller_array_export_as_shelx_hklf(intensities, hkl_file_handle,
scale_range=(-9999., 9999.), normalise_if_format_overflow=True)
crystal_symm = intensities.crystal_symmetry()
wavelength = options.wavelength
if wavelength is None:
mtz_crystals = mtz_object.crystals()
wavelength = mtz_crystals[1].datasets()[0].wavelength()
print 'Experimental wavelength: %.3f Angstroms' % wavelength
unit_cell_dims = None
unit_cell_esds = None
cell_data = None
if options.cell:
with open(options.cell, 'r') as fh:
cell_data = json.load(fh)
solution = cell_data.get('solution_constrained', cell_data['solution_unconstrained'])
unit_cell_dims = tuple([
solution[dim]['mean'] for dim in ['a', 'b', 'c', 'alpha', 'beta', 'gamma']
])
unit_cell_esds = tuple([
solution[dim]['population_standard_deviation'] for dim in ['a', 'b', 'c', 'alpha', 'beta', 'gamma']
])
cb_op = crystal_symm.change_of_basis_op_to_reference_setting()
if cb_op.c().r().as_hkl() == 'h,k,l':
print 'Change of basis to reference setting: %s' % cb_op
crystal_symm = crystal_symm.change_basis(cb_op)
if str(cb_op) != "a,b,c":
unit_cell_dims = None
unit_cell_esds = None
# Would need to apply operation to cell errors, too. Need a test case for this
# crystal_symm.show_summary()
xray_structure = structure(crystal_symmetry=crystal_symm)
if compound:
result = parse_compound(compound)
for element in result:
xray_structure.add_scatterer(scatterer(label=element,
occupancy=result[element]))
open('%s.ins' % prefix, 'w').write(''.join(
writer.generator(xray_structure,
wavelength=wavelength,
full_matrix_least_squares_cycles=0,
title=prefix,
unit_cell_esds=unit_cell_esds)))
generate_cif(prefix=prefix, unit_cell_data=cell_data, wavelength=wavelength, structure=xray_structure)
def _scale_finish_export_shelxt(self):
'''Read hklin (unmerged reflection file) and generate SHELXT input file
and HKL file'''
from iotbx.reflection_file_reader import any_reflection_file
from iotbx.shelx import writer
from iotbx.shelx.hklf import miller_array_export_as_shelx_hklf
from cctbx.xray.structure import structure
from cctbx.xray import scatterer
for wavelength_name in self._scalr_scaled_refl_files.keys():
prefix = wavelength_name
if len(self._scalr_scaled_refl_files.keys()) == 1:
prefix = 'shelxt'
prefixpath = os.path.join(self.get_working_directory(), prefix)
mtz_unmerged = self._scalr_scaled_reflection_files['mtz_unmerged'][
wavelength_name]
reader = any_reflection_file(mtz_unmerged)
intensities = [
ma for ma in reader.as_miller_arrays(merge_equivalents=False)
if ma.info().labels == ['I', 'SIGI']
][0]
# FIXME do I need to reindex to a conventional setting here
indices = reader.file_content(
).extract_original_index_miller_indices()
intensities = intensities.customized_copy(indices=indices,
info=intensities.info())
with open('%s.hkl' % prefixpath, 'wb') as hkl_file_handle:
# limit values to 4 digits (before decimal point), as this is what shelxt
# writes in its output files, and shelxl seems to read. ShelXL apparently
# does not read values >9999 properly
miller_array_export_as_shelx_hklf(
intensities,
hkl_file_handle,
scale_range=(-9999., 9999.),
normalise_if_format_overflow=True)
crystal_symm = intensities.crystal_symmetry()
unit_cell_dims = self._scalr_cell
unit_cell_esds = self._scalr_cell_esd
cb_op = crystal_symm.change_of_basis_op_to_reference_setting()
if cb_op.c().r().as_hkl() == 'h,k,l':
print('Change of basis to reference setting: %s' % cb_op)
crystal_symm = crystal_symm.change_basis(cb_op)
if str(cb_op) != "a,b,c":
unit_cell_dims = None
unit_cell_esds = None
# Would need to apply operation to cell errors, too. Need a test case for this
# crystal_symm.show_summary()
xray_structure = structure(crystal_symmetry=crystal_symm)
compound = 'CNOH'
if compound:
from xia2.command_line.to_shelx import parse_compound
result = parse_compound(compound)
for element in result:
xray_structure.add_scatterer(
scatterer(label=element, occupancy=result[element]))
wavelength = self._scalr_xcrystal.get_xwavelength(
wavelength_name).get_wavelength()
with open('%s.ins' % prefixpath, 'w') as insfile:
insfile.write(''.join(
writer.generator(xray_structure,
wavelength=wavelength,
full_matrix_least_squares_cycles=0,
title=prefix,
unit_cell_dims=unit_cell_dims,
unit_cell_esds=unit_cell_esds)))
FileHandler.record_data_file('%s.ins' % prefixpath)
FileHandler.record_data_file('%s.hkl' % prefixpath)
