def get_HKL(self,cursor):
name = self.db_experiment_tag
query = '''SELECT H,K,L,%s_hkls.hkl_id from %s_hkls,%s_isoforms WHERE %s_hkls.isoforms_isoform_id = %s_isoforms.isoform_id AND %s_isoforms.name = "%s"'''%(
name, name, name, name, name, name, self.params["identified_isoform"])
cursor.execute(query)
ALL = cursor.fetchall()
indices = flex.miller_index([(a[0],a[1],a[2]) for a in ALL])
miller_id = flex.int([a[3] for a in ALL])
self.miller_set = mset(crystal_symmetry=self.params["observations"][0].crystal_symmetry(), indices=indices)
self.miller_set_id = miller_id
# might have to change this to isoform_id next iteration
cursor.execute('SELECT isoform_id FROM %s_isoforms WHERE name = "%s"'%(
name, self.params["identified_isoform"]))
self.isoform_id = cursor.fetchall()[0][0]
return indices,miller_id
def application(params, loop=True):
from cxi_xdr_xes.cftbx.cspad_ana import db as cxidb
dbobj = cxidb.dbconnect(params.db.host, params.db.name, params.db.user,
params.db.password)
cursor = dbobj.cursor()
PM = progress_manager(params, cursor)
PM.setup_isoforms(cursor)
PM.setup_runtags(cursor)
isoforms = PM.isoforms
del dbobj
while 1:
dbobj = cxidb.dbconnect(params.db.host, params.db.name, params.db.user,
params.db.password)
cursor = dbobj.cursor()
results = {}
print "Looking for data..."
for tag in params.run_tags.split(','):
for isoform in isoforms:
M = PM.get_HKL(cursor, isoform=isoform, run_tags=tag)
cell = isoforms[isoform]['cell']
miller_set = mset(
anomalous_flag=False,
crystal_symmetry=symmetry(
unit_cell=cell,
space_group_symbol=isoforms[isoform]['lookup_symbol']),
indices=M)
miller_set.show_comprehensive_summary()
miller_set.setup_binner(d_min=params.resolution,
n_bins=params.n_bins)
given = miller_set.binner().counts_given()
ccomplete = miller_set.binner().counts_complete()
for i_bin in miller_set.binner().range_used():
sel = miller_set.binner().selection(i_bin)
self_sel = miller_set.select(sel)
d_max, d_min = self_sel.d_max_min()
compl = self_sel.completeness(d_max=d_max)
n_ref = sel.count(True)
if ccomplete[i_bin] == 0.:
multiplicity = 0.
else:
res_highest = d_min
multiplicity = given[i_bin] / ccomplete[i_bin]
d_range = miller_set.binner().bin_legend(
i_bin=i_bin, show_bin_number=False, show_counts=True)
fmt = "%3d: %-24s %4.2f %6d mult=%4.2f"
print fmt % (i_bin, d_range, compl, n_ref, multiplicity)
print
if len(tag) > 0:
key = "%s %s" % (tag, isoform)
else:
key = isoform
given_used = flex.int(given).select(
flex.size_t(miller_set.binner().range_used()))
ccomplete_used = flex.int(ccomplete).select(
flex.size_t(miller_set.binner().range_used()))
results[key] = dict(multiplicity=flex.sum(given_used) /
flex.sum(ccomplete_used),
completeness=miller_set.completeness(),
multiplicity_highest=multiplicity,
completeness_highest=compl,
resolution_highest=res_highest)
del dbobj
if not loop:
return results
time.sleep(10)
def run_job():
from cctbx.crystal import symmetry
from cctbx.uctbx import unit_cell
from cctbx.sgtbx import space_group
sym = symmetry(unit_cell=unit_cell((78.68, 78.68, 265.33, 90, 90, 120)),
space_group="P6522")
millers = flex.miller_index()
for ref, exp in from_files():
idx = ref["miller_index"]
millers.extend(idx)
from cctbx.miller import set as mset, array
MM = mset(crystal_symmetry=sym, anomalous_flag=True,
indices=millers) # make miller set from all obs
MA = MM.map_to_asu() # map to asu
#dval = MA.unit_cell().d(MA.indices())
#order = flex.sort_permutation(dval, reverse=True)
#MAsort = MA.select(order)
#for item in MAsort.indices(): print(item)
#print (MAsort.indices().size(), len(set(MAsort.indices())))
#ZZ = MAsort.array(data=flex.double(MAsort.size(),1.))
MAobs = MA.array(data=flex.double(MA.size(), 1.))
ME = MAobs.merge_equivalents()
new_red = ME.redundancies(
) # all observations with multiplicity of measurement
z = new_red.complete_array(
d_max=100000., new_data_value=0) # fill in missing observations
dval = z.unit_cell().d(z.indices())
order = flex.sort_permutation(dval, reverse=True)
MAsort = z.select(order)
print("d_min is %.3f" % (flex.min(MA.d_spacings().data())))
print("there are altogether %d measurements" % MA.indices().size())
print("this is uniqueness : %d measurements" % new_red.indices().size())
print("total in asymmetric unit %d" % MAsort.indices().size())
print("maximum multiplicity is %d" % (flex.max(MAsort.data())))
print("the mean multiplicity is %.2f" %
(flex.mean(MAsort.data().as_double())))
zeroes = (MAsort.data() == 0).count(True)
print("unmeasured %d or %.2f%% of total" %
(zeroes, 100. * zeroes / MAsort.indices().size()))
MAd = MA.select(MA.d_spacings().data() > 1.63)
new_redd = new_red.select(new_red.d_spacings().data() > 1.63)
MAsortd = MAsort.select(MAsort.d_spacings().data() > 1.63)
print("Cut, there are altogether %d measurements" % MAd.indices().size())
print("Cut, this is uniqueness : %d measurements" %
new_redd.indices().size())
print("Cut, total in asymmetric unit %d" % MAsortd.indices().size())
print("Cut, maximum multiplicity is %d" % (flex.max(MAsortd.data())))
print("Cut, the mean multiplicity is %.2f" %
(flex.mean(MAsortd.data().as_double())))
zeroes = (MAsortd.data() == 0).count(True)
print("Cut, unmeasured %d or %.2f%% of total" %
(zeroes, 100. * zeroes / MAsortd.indices().size()))
if True:
from matplotlib import pyplot as plt
plt.plot(range(MAsort.indices().size()), MAsort.data(), "r.")
plt.plot(range(MAsortd.indices().size()),
MAsortd.data().as_double() - 0.1, "b.")
plt.title("Redundant measurement vs. Bragg spot position")
plt.xlabel("Spots ordered by increasing Bragg angle →")
plt.ylabel("Multiplicity of strong spot obs (Group A)")
plt.show()
def application(params, loop = True):
from cxi_xdr_xes.cftbx.cspad_ana import db as cxidb
dbobj = cxidb.dbconnect(params.db.host, params.db.name, params.db.user, params.db.password)
cursor = dbobj.cursor()
PM = progress_manager(params, cursor)
PM.setup_isoforms(cursor)
PM.setup_runtags(cursor)
isoforms = PM.isoforms
del dbobj
while 1:
dbobj = cxidb.dbconnect(params.db.host, params.db.name, params.db.user, params.db.password)
cursor = dbobj.cursor()
results = {}
print "Looking for data..."
for tag in params.run_tags.split(','):
for isoform in isoforms:
M = PM.get_HKL(cursor,isoform=isoform,run_tags=tag)
cell = isoforms[isoform]['cell']
miller_set = mset(anomalous_flag = False, crystal_symmetry=symmetry(unit_cell=cell, space_group_symbol=isoforms[isoform]['lookup_symbol']), indices=M)
miller_set.show_comprehensive_summary()
miller_set.setup_binner(d_min=params.resolution, n_bins=params.n_bins)
given = miller_set.binner().counts_given()
ccomplete = miller_set.binner().counts_complete()
for i_bin in miller_set.binner().range_used():
sel = miller_set.binner().selection(i_bin)
self_sel = miller_set.select(sel)
d_max,d_min = self_sel.d_max_min()
compl = self_sel.completeness(d_max = d_max)
n_ref = sel.count(True)
if ccomplete[i_bin] == 0.:
multiplicity = 0.
else:
res_highest = d_min
multiplicity = given[i_bin]/ccomplete[i_bin]
d_range = miller_set.binner().bin_legend(
i_bin = i_bin, show_bin_number = False, show_counts = True)
fmt = "%3d: %-24s %4.2f %6d mult=%4.2f"
print fmt % (i_bin,d_range,compl,n_ref,
multiplicity)
print
if len(tag) > 0:
key = "%s %s"%(tag, isoform)
else:
key = isoform
given_used = flex.int(given).select(flex.size_t(miller_set.binner().range_used()))
ccomplete_used = flex.int(ccomplete).select(flex.size_t(miller_set.binner().range_used()))
results[key] = dict(
multiplicity = flex.sum(given_used)/flex.sum(ccomplete_used),
completeness = miller_set.completeness(),
multiplicity_highest = multiplicity,
completeness_highest = compl,
resolution_highest = res_highest
)
del dbobj
if not loop:
return results
time.sleep(10)