def __init__(self,
mtz_object,
custom_cif_labels_dict=None,
log=None,
test_flag_value=None):
self.cif_blocks = {'xray': None, 'neutron': None}
if log is None: log = sys.stdout
miller_arrays = mtz_object.as_miller_arrays()
miller_arrays_as_cif_block = None
input_observations_xray = None
input_observations_neutron = None
r_free_xray = None
r_free_neutron = None
f_obs_filtered_xray = None
f_obs_filtered_neutron = None
mtz_to_cif_labels_dict = {}
mtz_to_cif_labels_dict.update(phenix_to_cif_labels_dict)
mtz_to_cif_labels_dict.update(ccp4_to_cif_labels_dict)
if custom_cif_labels_dict is not None:
mtz_to_cif_labels_dict.update(custom_cif_labels_dict)
unknown_mtz_labels = []
for array in miller_arrays:
labels = array.info().labels
label = labels[0]
if reflection_file_utils.looks_like_r_free_flags_info(
array.info()):
if "(+)" in label:
array = array.average_bijvoet_mates()
labels = [label.replace("(+)", "")]
if label.endswith(("neutron", "_N")):
r_free_neutron = array
else:
r_free_xray = array
continue # deal with these later
elif label.startswith("F-obs-filtered"):
if label.endswith(("neutron", "_N")):
f_obs_filtered_neutron = array
else:
f_obs_filtered_xray = array
elif label.startswith("F-obs") or label.startswith("I-obs"):
if label.strip("(+)").endswith(("neutron", "_N")):
input_observations_neutron = array
else:
input_observations_xray = array
#elif label.startswith("R-free-flags"):
column_names = []
for mtz_label in labels:
cif_label = mtz_to_cif_label(mtz_to_cif_labels_dict, mtz_label)
column_names.append(cif_label)
if column_names.count(None) > 0:
# I don't know what to do with this array
for i, mtz_label in enumerate(labels):
if column_names[i] is None:
unknown_mtz_labels.append(mtz_label)
continue
assert column_names.count(None) == 0
if labels[0].strip("(+)").endswith(("neutron", "_N")):
data_type = "neutron"
else:
data_type = "xray"
if column_names[0].startswith(
("_refln.F_meas", "_refln.F_squared_meas", "_refln.pdbx_F_",
"_refln.pdbx_I_")):
if data_type == "neutron":
input_observations_neutron = array
else:
input_observations_xray = array
if self.cif_blocks.get(data_type) is None:
self.cif_blocks[
data_type] = iotbx.cif.miller_arrays_as_cif_block(
array=array, column_names=column_names, format="mmcif")
else:
self.cif_blocks[data_type].add_miller_array(
array, column_names=column_names)
if len(unknown_mtz_labels):
print >> log, "Warning: Unknown mtz label%s: %s" % (plural_s(
len(unknown_mtz_labels))[1], ", ".join(unknown_mtz_labels))
print >> log, " Use mtz_labels and cif_labels keywords to provide translation for custom labels."
data_types = set(["xray"])
if self.cif_blocks['neutron'] is not None:
data_types.add("neutron")
if input_observations_xray is None and f_obs_filtered_xray is not None:
self.cif_blocks["xray"].add_miller_array(
array=f_obs_filtered_xray,
column_names=('_refln.F_meas_au', '_refln.F_meas_sigma_au'))
if input_observations_neutron is None and f_obs_filtered_neutron is not None:
self.cif_blocks["neutron"].add_miller_array(
array=f_obs_filtered_neutron,
column_names=('_refln.F_meas_au', '_refln.F_meas_sigma_au'))
for data_type in data_types:
if data_type == "xray":
r_free = r_free_xray
input_obs = input_observations_xray
f_obs_filtered = f_obs_filtered_xray
if (self.cif_blocks["xray"] is None and r_free_xray is not None
and self.cif_blocks["neutron"] is not None
and r_free_neutron is None):
r_free_neutron = r_free_xray
elif data_type == "neutron":
r_free = r_free_neutron
input_obs = input_observations_neutron
f_obs_filtered = f_obs_filtered_neutron
if self.cif_blocks[data_type] is not None and r_free is not None:
self.cif_blocks[data_type].add_miller_array(
array=r_free, column_name='_refln.pdbx_r_free_flag')
if input_obs is None or r_free is None: continue
if (test_flag_value is None):
test_flag_value = reflection_file_utils.guess_r_free_flag_value(
miller_array=r_free)
assert (test_flag_value is not None)
refln_status = r_free.array(
data=flex.std_string(r_free.size(), "."))
input_obs_non_anom = input_obs.average_bijvoet_mates()
match = r_free.match_indices(input_obs_non_anom)
refln_status.data().set_selected(match.pair_selection(0), "o")
refln_status.data().set_selected(r_free.data() == test_flag_value,
"f")
if f_obs_filtered is not None:
f_obs_filtered_non_anom = f_obs_filtered.average_bijvoet_mates(
)
match = r_free.match_indices(f_obs_filtered_non_anom)
refln_status.data().set_selected(match.single_selection(0),
"<") # XXX
self.cif_blocks[data_type].add_miller_array(
array=refln_status, column_name="_refln.status")
def __init__(self, mtz_object, custom_cif_labels_dict=None, log=None,
test_flag_value=None):
self.cif_blocks = {
'xray': None,
'neutron': None
}
if log is None: log = sys.stdout
miller_arrays = mtz_object.as_miller_arrays()
miller_arrays_as_cif_block = None
input_observations_xray = None
input_observations_neutron = None
r_free_xray = None
r_free_neutron = None
f_obs_filtered_xray = None
f_obs_filtered_neutron = None
mtz_to_cif_labels_dict = {}
mtz_to_cif_labels_dict.update(phenix_to_cif_labels_dict)
mtz_to_cif_labels_dict.update(ccp4_to_cif_labels_dict)
if custom_cif_labels_dict is not None:
mtz_to_cif_labels_dict.update(custom_cif_labels_dict)
unknown_mtz_labels = []
for array in miller_arrays:
labels = array.info().labels
label = labels[0]
if reflection_file_utils.looks_like_r_free_flags_info(array.info()):
if "(+)" in label:
array = array.average_bijvoet_mates()
labels = [label.replace("(+)", "")]
if label.endswith(("neutron", "_N")):
r_free_neutron = array
else:
r_free_xray = array
continue # deal with these later
elif label.startswith("F-obs-filtered"):
if label.endswith(("neutron", "_N")):
f_obs_filtered_neutron = array
else:
f_obs_filtered_xray = array
elif label.startswith("F-obs") or label.startswith("I-obs"):
if label.strip("(+)").endswith(("neutron", "_N")):
input_observations_neutron = array
else:
input_observations_xray = array
#elif label.startswith("R-free-flags"):
column_names = []
for mtz_label in labels:
cif_label = mtz_to_cif_label(mtz_to_cif_labels_dict, mtz_label)
column_names.append(cif_label)
if column_names.count(None) > 0:
# I don't know what to do with this array
for i, mtz_label in enumerate(labels):
if column_names[i] is None:
unknown_mtz_labels.append(mtz_label)
continue
assert column_names.count(None) == 0
if labels[0].strip("(+)").endswith(("neutron", "_N")):
data_type = "neutron"
else:
data_type = "xray"
if column_names[0].startswith(("_refln.F_meas",
"_refln.F_squared_meas",
"_refln.pdbx_F_",
"_refln.pdbx_I_")):
if data_type == "neutron":
input_observations_neutron = array
else:
input_observations_xray = array
if self.cif_blocks.get(data_type) is None:
self.cif_blocks[data_type] = iotbx.cif.miller_arrays_as_cif_block(
array=array, column_names=column_names, format="mmcif")
else:
self.cif_blocks[data_type].add_miller_array(array, column_names=column_names)
if len(unknown_mtz_labels):
print >> log, "Warning: Unknown mtz label%s: %s" %(
plural_s(len(unknown_mtz_labels))[1], ", ".join(unknown_mtz_labels))
print >> log, " Use mtz_labels and cif_labels keywords to provide translation for custom labels."
data_types = set(["xray"])
if self.cif_blocks['neutron'] is not None:
data_types.add("neutron")
if input_observations_xray is None and f_obs_filtered_xray is not None:
self.cif_blocks["xray"].add_miller_array(
array=f_obs_filtered_xray,
column_names=('_refln.F_meas_au','_refln.F_meas_sigma_au'))
if input_observations_neutron is None and f_obs_filtered_neutron is not None:
self.cif_blocks["neutron"].add_miller_array(
array=f_obs_filtered_neutron,
column_names=('_refln.F_meas_au','_refln.F_meas_sigma_au'))
for data_type in data_types:
if data_type == "xray":
r_free = r_free_xray
input_obs = input_observations_xray
f_obs_filtered = f_obs_filtered_xray
if (self.cif_blocks["xray"] is None and r_free_xray is not None and
self.cif_blocks["neutron"] is not None and r_free_neutron is None):
r_free_neutron = r_free_xray
elif data_type == "neutron":
r_free = r_free_neutron
input_obs = input_observations_neutron
f_obs_filtered = f_obs_filtered_neutron
if self.cif_blocks[data_type] is not None and r_free is not None:
self.cif_blocks[data_type].add_miller_array(
array=r_free, column_name='_refln.pdbx_r_free_flag')
if input_obs is None or r_free is None: continue
if (test_flag_value is None) :
test_flag_value = reflection_file_utils.guess_r_free_flag_value(
miller_array=r_free)
assert (test_flag_value is not None)
refln_status = r_free.array(data=flex.std_string(r_free.size(), "."))
input_obs_non_anom = input_obs.average_bijvoet_mates()
match = r_free.match_indices(input_obs_non_anom)
refln_status.data().set_selected(match.pair_selection(0), "o")
refln_status.data().set_selected(r_free.data() == test_flag_value, "f")
if f_obs_filtered is not None:
f_obs_filtered_non_anom = f_obs_filtered.average_bijvoet_mates()
match = r_free.match_indices(f_obs_filtered_non_anom)
refln_status.data().set_selected(match.single_selection(0), "<") # XXX
self.cif_blocks[data_type].add_miller_array(
array=refln_status, column_name="_refln.status")
def exercise_get_r_free_flags():
crystal_symmetry = crystal.symmetry(
unit_cell=(30,31,32,85,95,100),
space_group_symbol="P 1")
miller_set = miller.build_set(
crystal_symmetry=crystal_symmetry,
anomalous_flag=False,
d_min=3)
n = miller_set.indices().size()
exercise_flag_arrays = []
exercise_flag_arrays.append(
flex.int(list(flex.random_permutation(size=n)%10)))
exercise_flag_arrays.append(flex.int(xrange(n)))
exercise_flag_arrays.append(flex.int(n, 0))
for style in ["ccp4", "cns", "shelx", "bool"]:
for i_exercise,exercise_flag_array in enumerate(exercise_flag_arrays):
for reversed in [False, True]:
if (style == "ccp4"):
if (reversed): break
data = exercise_flag_array
test_flag_value = 3
else:
if (not reversed):
data = (exercise_flag_array == 0)
test_flag_value = True
else:
data = (exercise_flag_array != 0)
test_flag_value = False
if (style == "cns"):
data = data.as_int()
test_flag_value = int(test_flag_value)
elif (style == "shelx"):
data = -data.as_int()
data.set_selected((data == 0), 1)
if (not reversed): test_flag_value = -1
else: test_flag_value = 1
input_array = miller_set.array(data=data)
mtz_dataset = input_array.as_mtz_dataset(
column_root_label="FreeRflags")
mtz_dataset.mtz_object().write("tmp.mtz")
reflection_files = [reflection_file_reader.any_reflection_file(
file_name="tmp.mtz")]
err = StringIO()
reflection_file_srv = reflection_file_server(
crystal_symmetry=crystal_symmetry,
force_symmetry=True,
reflection_files=reflection_files,
err=err)
for trial_test_flag_value in [None, test_flag_value]:
for trial_label in [None, "free", "foo"]:
try:
r_free_flags, actual_test_flag_value = \
reflection_file_srv.get_r_free_flags(
file_name=None,
label=trial_label,
test_flag_value=trial_test_flag_value,
disable_suitability_test=False,
parameter_scope="r_free_flags")
except Sorry, e:
if (trial_label != "foo"):
assert i_exercise > 0
if (trial_label is None):
assert str(e) == """\
No array of R-free flags found.
For manual selection define:
r_free_flags.label
r_free_flags.test_flag_value
r_free_flags.disable_suitability_test=True"""
else:
assert str(e) == \
"Not a suitable array of R-free flags:" \
+ " r_free_flags.label=free\n" \
+ "To override the suitability test define:" \
+ " r_free_flags.disable_suitability_test=True"
else:
assert str(e) == "No matching array: r_free_flags.label=foo"
if (i_exercise == 0):
assert err.getvalue() == """\
No matching array: r_free_flags.label=foo
Possible choices:
tmp.mtz:FreeRflags
Please use r_free_flags.label
to specify an unambiguous substring of the target label.
"""
else:
assert err.getvalue() == """\
No matching array: r_free_flags.label=foo
"""
err = reflection_file_srv.err = StringIO()
else:
assert i_exercise == 0
actual_test_flag_value_2 = guess_r_free_flag_value(
miller_array=r_free_flags,
test_flag_value=trial_test_flag_value)
assert (actual_test_flag_value_2 == actual_test_flag_value)
def exercise_get_r_free_flags():
crystal_symmetry = crystal.symmetry(
unit_cell=(30,31,32,85,95,100),
space_group_symbol="P 1")
miller_set = miller.build_set(
crystal_symmetry=crystal_symmetry,
anomalous_flag=False,
d_min=3)
n = miller_set.indices().size()
exercise_flag_arrays = []
exercise_flag_arrays.append(
flex.int(list(flex.random_permutation(size=n)%10)))
exercise_flag_arrays.append(flex.int(range(n)))
exercise_flag_arrays.append(flex.int(n, 0))
for style in ["ccp4", "cns", "shelx", "bool"]:
for i_exercise,exercise_flag_array in enumerate(exercise_flag_arrays):
for reversed in [False, True]:
if (style == "ccp4"):
if (reversed): break
data = exercise_flag_array
test_flag_value = 3
else:
if (not reversed):
data = (exercise_flag_array == 0)
test_flag_value = True
else:
data = (exercise_flag_array != 0)
test_flag_value = False
if (style == "cns"):
data = data.as_int()
test_flag_value = int(test_flag_value)
elif (style == "shelx"):
data = -data.as_int()
data.set_selected((data == 0), 1)
if (not reversed): test_flag_value = -1
else: test_flag_value = 1
input_array = miller_set.array(data=data)
mtz_dataset = input_array.as_mtz_dataset(
column_root_label="FreeRflags")
mtz_dataset.mtz_object().write("tmp.mtz")
reflection_files = [reflection_file_reader.any_reflection_file(
file_name="tmp.mtz")]
err = StringIO()
reflection_file_srv = reflection_file_server(
crystal_symmetry=crystal_symmetry,
force_symmetry=True,
reflection_files=reflection_files,
err=err)
for trial_test_flag_value in [None, test_flag_value]:
for trial_label in [None, "free", "foo"]:
try:
r_free_flags, actual_test_flag_value = \
reflection_file_srv.get_r_free_flags(
file_name=None,
label=trial_label,
test_flag_value=trial_test_flag_value,
disable_suitability_test=False,
parameter_scope="r_free_flags")
except Sorry as e:
if (trial_label != "foo"):
assert i_exercise > 0
if (trial_label is None):
assert str(e) == """\
No array of R-free flags found.
For manual selection define:
r_free_flags.label
r_free_flags.test_flag_value
r_free_flags.disable_suitability_test=True"""
else:
assert str(e) == \
"Not a suitable array of R-free flags:" \
+ " r_free_flags.label=free\n" \
+ "To override the suitability test define:" \
+ " r_free_flags.disable_suitability_test=True"
else:
assert str(e) == "No matching array: r_free_flags.label=foo"
if (i_exercise == 0):
assert err.getvalue() == """\
No matching array: r_free_flags.label=foo
Possible choices:
tmp.mtz:FreeRflags
Please use r_free_flags.label
to specify an unambiguous substring of the target label.
"""
else:
assert err.getvalue() == """\
No matching array: r_free_flags.label=foo
"""
err = reflection_file_srv.err = StringIO()
else:
assert i_exercise == 0
actual_test_flag_value_2 = guess_r_free_flag_value(
miller_array=r_free_flags,
test_flag_value=trial_test_flag_value)
assert (actual_test_flag_value_2 == actual_test_flag_value)
for second_label in ["test", "foo"]:
input_array = miller_set.array(data=exercise_flag_arrays[0])
mtz_dataset = input_array.as_mtz_dataset(
column_root_label="FreeRflags")
mtz_dataset.add_miller_array(
miller_array=input_array,
column_root_label=second_label)
mtz_dataset.mtz_object().write("tmp.mtz")
reflection_files = [reflection_file_reader.any_reflection_file(
file_name="tmp.mtz")]
err = StringIO()
reflection_file_srv = reflection_file_server(
crystal_symmetry=crystal_symmetry,
force_symmetry=True,
reflection_files=reflection_files,
err=err)
try:
reflection_file_srv.get_r_free_flags(
file_name=None,
label=None,
test_flag_value=None,
disable_suitability_test=False,
parameter_scope="r_free_flags")
except Sorry as e:
assert str(e)=="Multiple equally suitable arrays of R-free flags found."
assert err.getvalue() == """\
Multiple equally suitable arrays of R-free flags found.
Possible choices:
tmp.mtz:FreeRflags
tmp.mtz:test
Please use r_free_flags.label
to specify an unambiguous substring of the target label.
"""
err = reflection_file_srv.err = StringIO()
else:
assert str(r_free_flags.info()) == "tmp.mtz:FreeRflags"
r_free_flags, actual_test_flag_value = \
reflection_file_srv.get_r_free_flags(
file_name=None,
label="FreeRflags",
test_flag_value=3,
disable_suitability_test=True,
parameter_scope="r_free_flags")
assert r_free_flags.info().label_string() == "FreeRflags"
assert actual_test_flag_value == 3
for label,test_flag_value in [(None,3), ("FreeRflags",None)]:
try:
reflection_file_srv.get_r_free_flags(
file_name=None,
label=label,
test_flag_value=test_flag_value,
disable_suitability_test=True,
parameter_scope="r_free_flags")
except Sorry as e:
assert str(e) == "r_free_flags.disable_suitability_test=True:" \
" Suitability test for R-free flags can only be disabled if both" \
" r_free_flags.label and r_free_flags.test_flag_value are defined."
else: raise Exception_expected
# test corrupted R-free flags
r_free_flags = miller_set.generate_r_free_flags()
int_flags = r_free_flags.data().as_int()
int_flags[100] = 10000000
r_free_flags = r_free_flags.customized_copy(data=int_flags)
mtz_dataset = r_free_flags.as_mtz_dataset(
column_root_label="TEST")
mtz_dataset.mtz_object().write("tmp.mtz")
reflection_files = [reflection_file_reader.any_reflection_file(
file_name="tmp.mtz")]
err = StringIO()
reflection_file_srv = reflection_file_server(
crystal_symmetry=crystal_symmetry,
force_symmetry=True,
reflection_files=reflection_files,
err=err)
flags, value = reflection_file_srv.get_r_free_flags(
file_name=None,
label=None,
test_flag_value=None,
disable_suitability_test=False,
parameter_scope="r_free_flags")
assert (value == 1)
