def run(self):
from dials.algorithms.image.fill_holes import simple_fill
from scitbx.array_family import flex
from random import randint
from math import sqrt
import sys
mask = flex.bool(flex.grid(100, 100), True)
data = flex.double(flex.grid(100, 100), True)
for j in range(100):
for i in range(100):
data[j,i] = 10 + j * 0.01 + i * 0.01
if sqrt((j - 50)**2 + (i - 50)**2) <= 10.5:
mask[j,i] = False
data[j,i] = 0
result = simple_fill(data, mask)
known = data.as_1d().select(mask.as_1d())
filled = result.as_1d().select(mask.as_1d() == False)
assert flex.max(filled) <= flex.max(known)
assert flex.min(filled) >= flex.min(known)
# Test passed
print 'OK'
def test_cma_es_rosebrock_n(M=10):
def funct(x,y):
result = 0
for xx,yy in zip(x,y):
result+=100.0*((yy-xx*xx)**2.0) + (1-xx)**2.0
return result
N=M*2
x = flex.double(N,10.0)
sd = flex.double(N,3.0)
m = cma_es(N,x,sd)
while ( not m.converged() ):
# sample population
p = m.sample_population()
pop_size = p.accessor().all()[0]
# update objective function
v = flex.double(pop_size)
for ii in range(pop_size):
vector = p[(ii*N):(ii*N + N)]
x = vector[0:M]
y = vector[M:]
v[ii] = funct(x,y)
m.update_distribution(v)
print list(m.get_result())
print flex.min(v)
print
x_final = m.get_result()
print list(x_final)
def exercise_reference_impl_long(n_dynamics_steps, out):
sim = fmri.simulation()
e_tots = flex.double([sim.e_tot])
print >> out, "i_step, [e_pot, e_kin_ang, e_kin_lin, e_kin, e_tot]"
def show(i_step):
print >> out, i_step, [sim.e_pot, sim.e_kin_ang, sim.e_kin_lin, sim.e_kin, sim.e_tot]
out.flush()
n_show = max(1, n_dynamics_steps // 10)
for i_step in xrange(n_dynamics_steps):
sim.dynamics_step(delta_t=0.001)
e_tots.append(sim.e_tot)
if i_step % n_show == 0:
show(i_step)
show(n_dynamics_steps)
print >> out
print >> out, "number of dynamics steps:", n_dynamics_steps
print >> out, "e_tot start:", e_tots[0]
print >> out, " final:", e_tots[-1]
print >> out, " min:", flex.min(e_tots)
print >> out, " max:", flex.max(e_tots)
print >> out, " max-min:", flex.max(e_tots) - flex.min(e_tots)
print >> out
out.flush()
def remove_common_isotropic_adp(self): xrs = self.xray_structure b_iso_min = flex.min(xrs.extract_u_iso_or_u_equiv()*adptbx.u_as_b(1)) self.b_overall = b_iso_min print >> self.log, "Max B subtracted from atoms and used to sharpen map:", b_iso_min xrs.shift_us(b_shift=-b_iso_min) b_iso_min = flex.min(xrs.extract_u_iso_or_u_equiv()*adptbx.u_as_b(1)) assert approx_equal(b_iso_min, 0, 1.e-3)
def run(args):
import libtbx.load_env
from dials.array_family import flex
from dials.util import log
from dials.util.version import dials_version
usage = "%s [options] experiment.json indexed.pickle" % \
libtbx.env.dispatcher_name
parser = OptionParser(
usage=usage,
phil=phil_scope,
read_reflections=True,
read_experiments=True,
check_format=False,
epilog=help_message)
params, options = parser.parse_args(show_diff_phil=True)
# Configure the logging
log.config(info=params.output.log, debug=params.output.debug_log)
logger.info(dials_version())
reflections = flatten_reflections(params.input.reflections)
experiments = flatten_experiments(params.input.experiments)
if len(reflections) == 0 or len(experiments) == 0:
parser.print_help()
return
assert(len(reflections) == 1)
assert(len(experiments) == 1)
experiment = experiments[0]
reflections = reflections[0]
# remove reflections with 0, 0, 0 index
zero = (reflections['miller_index'] == (0, 0, 0))
logger.info('Removing %d unindexed reflections' % zero.count(True))
reflections = reflections.select(~zero)
h, k, l = reflections['miller_index'].as_vec3_double().parts()
h = h.iround()
k = k.iround()
l = l.iround()
logger.info('Range on h: %d to %d' % (flex.min(h), flex.max(h)))
logger.info('Range on k: %d to %d' % (flex.min(k), flex.max(k)))
logger.info('Range on l: %d to %d' % (flex.min(l), flex.max(l)))
test_P1_crystal_indexing(reflections, experiment, params)
test_crystal_pointgroup_symmetry(reflections, experiment, params)
def tst_curve_interpolator():
x = flex.double( range(25) )/24.0
y = x*x
ip = curve_interpolator(0,2.0,200)
x_target = ip.target_x
y_ref = x_target*x_target
nx,ny,a,b = ip.interpolate(x,y)
count = 0
for xx in x_target:
if flex.max(x) >= xx:
count += 1
assert count==len(nx)
for yy,yyy in zip(ny,y_ref):
assert approx_equal(yy,yyy,eps=1e-3)
assert a[0]==0
assert a[1]==24
assert b[0]==0
assert b[1] in (99,100)
x = flex.double( range(5,23) )/24.0
y = x*x
ip = curve_interpolator(0,2.0,200)
nx,ny,a,b = ip.interpolate(x,y)
assert nx[0] >= flex.min(x)
assert nx[-1] <= flex.max(x)
y_ref= nx*nx
for yy,yyy in zip(ny,y_ref):
assert approx_equal(yy,yyy,eps=1e-3)
def inject(self,c=3.0):
mdelta = flex.max(self.mean - self.last_mean)
sdelta = flex.min(self.sigma)
if sdelta < self.inject_eps:
self.sigma = self.sigma + max(mdelta*c,c*self.inject_eps)
self.last_mean = self.mean.deep_copy()
self.last_sigma = self.sigma.deep_copy()
def box_iterator(self):
b = maptbx.boxes(
n_real = self.atom_map_asu.focus(),
fraction = self.box_size_as_fraction,
max_boxes= self.max_boxes,
log = self.log)
def get_wide_box(s,e): # define wide box: neutral + phased volumes
if(self.neutral_volume_box_cushion_width>0):
sh = self.neutral_volume_box_cushion_width
ss = [max(s[i]-sh,0) for i in [0,1,2]]
ee = [min(e[i]+sh,n_real_asu[i]) for i in [0,1,2]]
else: ss,ee = s,e
return ss,ee
n_real_asu = b.n_real
n_boxes = len(b.starts)
i_box = 0
for s,e in zip(b.starts, b.ends):
i_box+=1
sw,ew = get_wide_box(s=s,e=e)
fmodel_omit = self.omit_box(start=sw, end=ew)
r = fmodel_omit.r_work()
self.r.append(r) # for tests only
if(self.log):
print >> self.log, "r(curr,min,max,mean)=%6.4f %6.4f %6.4f %6.4f"%(r,
flex.min(self.r), flex.max(self.r), flex.mean(self.r)), i_box, n_boxes
omit_map_data = self.asu_map_from_fmodel(
fmodel=fmodel_omit, map_type=self.map_type)
maptbx.copy_box(
map_data_from = omit_map_data,
map_data_to = self.map_result_asu,
start = s,
end = e)
self.map_result_asu.reshape(self.acc_asu)
def prepare_maps(fofc, two_fofc, fem, fofc_cutoff=2, two_fofc_cutoff=0.5,
fem_cutoff=0.5, connectivity_cutoff=0.5, local_average=True):
"""
- This takes 3 maps: mFo-DFc, 2mFo-DFc and FEM and combines them into one map
that is most suitable for real-space refinement.
- Maps are the boxes extracted around region of interest from the whole unit
cell map.
- All maps are expected to be normalized by standard deviation (sigma-scaled)
BEFORE extracting the box. There is no way to assert it at this point.
- Map gridding equivalence is asserted.
"""
m1,m2,m3 = fofc, two_fofc, fem
# assert identical gridding
for m_ in [m1,m2,m3]:
for m__ in [m1,m2,m3]:
assert m_.all() == m__.all()
assert m_.focus() == m__.focus()
assert m_.origin() == m__.origin()
# binarize residual map
sel = m1 <= fofc_cutoff
mask = m1 .set_selected( sel, 0)
mask = mask.set_selected(~sel, 1)
del sel, m1
assert approx_equal([flex.max(mask), flex.min(mask)], [1,0])
def truncate_and_filter(m, cutoff, mask):
return m.set_selected(m<=cutoff, 0)*mask
# truncate and filter 2mFo-DFc map
m2 = truncate_and_filter(m2, two_fofc_cutoff, mask)
# truncate and filter FEM
m3 = truncate_and_filter(m3, fem_cutoff, mask)
del mask
# combined maps
def scale(m):
sd = m.sample_standard_deviation()
if(sd != 0): return m/sd
else: return m
m2 = scale(m2)
m3 = scale(m3)
m = (m2+m3)/2.
del m2, m3
m = scale(m)
# connectivity analysis
co = maptbx.connectivity(map_data=m, threshold=connectivity_cutoff)
v_max=-1.e+9
i_max=None
for i, v in enumerate(co.regions()):
if(i>0):
if(v>v_max):
v_max=v
i_max=i
mask2 = co.result()
selection = mask2==i_max
mask2 = mask2.set_selected(selection, 1)
mask2 = mask2.set_selected(~selection, 0)
assert mask2.count(1) == v_max
# final filter
m = m * mask2.as_double()
if(local_average):
maptbx.map_box_average(map_data=m, cutoff=0.5, index_span=1)
return m
def exercise_sim(out, n_dynamics_steps, delta_t, sim):
sim.check_d_pot_d_q()
e_pots = flex.double([sim.e_pot])
e_kins = flex.double([sim.e_kin])
for i_step in xrange(n_dynamics_steps):
sim.dynamics_step(delta_t=delta_t)
e_pots.append(sim.e_pot)
e_kins.append(sim.e_kin)
e_tots = e_pots + e_kins
sim.check_d_pot_d_q()
print >> out, "energy samples:", e_tots.size()
print >> out, "e_pot min, max:", min(e_pots), max(e_pots)
print >> out, "e_kin min, max:", min(e_kins), max(e_kins)
print >> out, "e_tot min, max:", min(e_tots), max(e_tots)
print >> out, "start e_tot:", e_tots[0]
print >> out, "final e_tot:", e_tots[-1]
ave = flex.sum(e_tots) / e_tots.size()
range = flex.max(e_tots) - flex.min(e_tots)
if (ave == 0): relative_range = 0
else: relative_range = range / ave
print >> out, "ave:", ave
print >> out, "range:", range
print >> out, "relative range:", relative_range
print >> out
out.flush()
if (out is sys.stdout):
f = open("tmp%02d.xy" % plot_number[0], "w")
for es in [e_pots, e_kins, e_tots]:
for e in es:
print >> f, e
print >> f, "&"
f.close()
plot_number[0] += 1
return relative_range
def exercise_tardy_model(out, n_dynamics_steps, delta_t, tardy_model):
tardy_model.check_d_e_pot_d_q()
e_pots = flex.double([tardy_model.e_pot()])
e_kins = flex.double([tardy_model.e_kin()])
for i_step in xrange(n_dynamics_steps):
tardy_model.dynamics_step(delta_t=delta_t)
e_pots.append(tardy_model.e_pot())
e_kins.append(tardy_model.e_kin())
e_tots = e_pots + e_kins
tardy_model.check_d_e_pot_d_q()
print >> out, "degrees of freedom:", tardy_model.degrees_of_freedom
print >> out, "energy samples:", e_tots.size()
print >> out, "e_pot min, max:", min(e_pots), max(e_pots)
print >> out, "e_kin min, max:", min(e_kins), max(e_kins)
print >> out, "e_tot min, max:", min(e_tots), max(e_tots)
print >> out, "start e_tot:", e_tots[0]
print >> out, "final e_tot:", e_tots[-1]
ave = flex.sum(e_tots) / e_tots.size()
range = flex.max(e_tots) - flex.min(e_tots)
if (ave == 0): relative_range = 0
else: relative_range = range / ave
print >> out, "ave:", ave
print >> out, "range:", range
print >> out, "relative range:", relative_range
print >> out
out.flush()
return relative_range
def __init__(self,rawdata,projection_vector,spotfinder_spot,verbose=False):
# projection vector is either the radial or azimuthal unit vector
# at a specific Bragg spot position
model_center = col((spotfinder_spot.ctr_mass_x(),spotfinder_spot.ctr_mass_y()))
px_x,px_y = project_2d_response_onto_line(projection_vector)
point_projections = flex.double()
pixel_values = flex.double()
for point in spotfinder_spot.bodypixels:
point_projection = (col((point.x,point.y)) - model_center).dot( projection_vector )
point_projections.append(point_projection)
pxval = rawdata[(point.x,point.y)]
if verbose:
print "point_projection",point_projection,
print "signal",pxval
pixel_values.append( pxval )
Lmin = flex.min(point_projections)
Lmax = flex.max(point_projections)
#print "Range %6.2f"%(Lmax-Lmin)
Rmin = round(Lmin-2.0,1)
Rmax = round(Lmax+2.0,1)
#print "Range %6.2f"%(Rmax-Rmin)
def histogram_bin (j) : return int(10.*(j-Rmin)) # bin units of 1/10 pixel
histo_x = flex.double((int(10*(Rmax-Rmin))))
histo_y = flex.double(len(histo_x))
for ihis in xrange(len(histo_x)): histo_x[ihis] = Rmin + 0.1*ihis
for ipp, point_projection in enumerate(point_projections):
value = pixel_values[ipp]
for isample in xrange(len(px_x)):
histo_y[int(10*(point_projection + px_x[isample] - Rmin))] += value * px_y[isample]
self.histo_x = histo_x
self.histo_y = histo_y
def blank_integrated_analysis(reflections, scan, phi_step, fractional_loss):
prf_sel = reflections.get_flags(reflections.flags.integrated_prf)
if prf_sel.count(True) > 0:
reflections = reflections.select(prf_sel)
intensities = reflections["intensity.prf.value"]
variances = reflections["intensity.prf.variance"]
else:
sum_sel = reflections.get_flags(reflections.flags.integrated_sum)
reflections = reflections.select(sum_sel)
intensities = reflections["intensity.sum.value"]
variances = reflections["intensity.sum.variance"]
i_sigi = intensities / flex.sqrt(variances)
xyz_px = reflections["xyzobs.px.value"]
x_px, y_px, z_px = xyz_px.parts()
phi = scan.get_angle_from_array_index(z_px)
osc = scan.get_oscillation()[1]
n_images_per_step = iceil(phi_step / osc)
phi_step = n_images_per_step * osc
phi_min = flex.min(phi)
phi_max = flex.max(phi)
n_steps = iceil((phi_max - phi_min) / phi_step)
hist = flex.histogram(z_px, n_slots=n_steps)
mean_i_sigi = flex.double()
for i, slot_info in enumerate(hist.slot_infos()):
sel = (z_px >= slot_info.low_cutoff) & (z_px < slot_info.high_cutoff)
if sel.count(True) == 0:
mean_i_sigi.append(0)
else:
mean_i_sigi.append(flex.mean(i_sigi.select(sel)))
fractional_mean_i_sigi = mean_i_sigi / flex.max(mean_i_sigi)
potential_blank_sel = mean_i_sigi <= (fractional_loss * flex.max(mean_i_sigi))
xmin, xmax = zip(*[(slot_info.low_cutoff, slot_info.high_cutoff) for slot_info in hist.slot_infos()])
d = {
"data": [
{
"x": list(hist.slot_centers()),
"y": list(mean_i_sigi),
"xlow": xmin,
"xhigh": xmax,
"blank": list(potential_blank_sel),
"type": "bar",
"name": "blank_counts_analysis",
}
],
"layout": {"xaxis": {"title": "z observed (images)"}, "yaxis": {"title": "Number of reflections"}, "bargap": 0},
}
blank_regions = blank_regions_from_sel(d["data"][0])
d["blank_regions"] = blank_regions
return d
def test1():
dials_regression = libtbx.env.find_in_repositories(
relative_path="dials_regression",
test=os.path.isdir)
data_dir = os.path.join(dials_regression, "centroid_test_data")
datablock_path = os.path.join(data_dir, "datablock.json")
# work in a temporary directory
cwd = os.path.abspath(os.curdir)
tmp_dir = open_tmp_directory(suffix="tst_rs_mapper")
os.chdir(tmp_dir)
cmd = 'dials.rs_mapper ' + datablock_path + ' map_file="junk.ccp4"'
result = easy_run.fully_buffered(command=cmd).raise_if_errors()
# load results
from iotbx import ccp4_map
from scitbx.array_family import flex
m = ccp4_map.map_reader(file_name="junk.ccp4")
assert len(m.data) == 7189057
assert approx_equal(m.header_min, -1.0)
assert approx_equal(flex.min(m.data), -1.0)
assert approx_equal(m.header_max, 2052.75)
assert approx_equal(flex.max(m.data), 2052.75)
assert approx_equal(m.header_mean, 0.018606403842568398)
assert approx_equal(flex.mean(m.data), 0.018606403842568398)
print "OK"
return
def apply_default_filter(database_dict, d_min, max_models_for_default_filter,
key = "high_resolution"):
database_dict = order_by_value(database_dict = database_dict, key = key)
values = flex.double()
for v in database_dict[key]: values.append(float(v))
diff = flex.abs(values-d_min)
min_val = flex.min(diff)
i_min_sel = (diff == min_val).iselection()
assert i_min_sel.size() > 0
i_min = i_min_sel[i_min_sel.size()//2]
i_l = max(0, i_min-max_models_for_default_filter//2)
i_r = min(values.size()-1, i_min+max_models_for_default_filter//2)
#
print "apply_default_filter:"
print " found data points dmin->higher =", abs(i_l-i_min)
print " found data points dmin->lower =", abs(i_r-i_min)
imm = min(abs(i_l-i_min), abs(i_r-i_min))
i_l, i_r = i_min-imm, i_min+imm
if (imm == 0) :
if (i_l == 0) :
i_r = 100
print " used data points dmin->higher =", 0
print " used data points dmin->lower =", i_r
elif (i_l == i_r == len(values) - 1) :
i_l -= 100
print " used data points dmin->higher =", i_l
print " used data points dmin->lower =", 0
else :
print " used data points dmin->higher =", imm
print " used data points dmin->lower =", imm
#
selection = flex.bool(values.size(), False)
for i in xrange(i_l,i_r): selection[i] = True
return select_dict(database_dict = database_dict, selection = selection)
def show(self):
b = self.bss_result
print >> self.log, " Statistics in resolution bins:"
#assert k_mask.size() == len(self.bin_selections)
fmt=" %7.5f %6.2f -%6.2f %5.1f %5d %-6s %-6s %-6s %6.3f %6.3f %8.2f %6.4f"
f_model = self.core.f_model.data()
print >> self.log, " s^2 Resolution Compl Nrefl k_mask k_iso k_ani <Fobs> R"
print >> self.log, " (A) (%) orig smooth average"
k_mask_bin_orig_ = str(None)
k_mask_bin_smooth_ = str(None)
k_mask_bin_approx_ = str(None)
for i_sel, cas in enumerate(self.cores_and_selections):
selection, core, selection_use, sel_work = cas
sel = sel_work
ss_ = self.ss_bin_values[i_sel][2]
if(b is not None and self.bss_result.k_mask_bin_orig is not None):
k_mask_bin_orig_ = "%6.4f"%self.bss_result.k_mask_bin_orig[i_sel]
if(b is not None and self.bss_result.k_mask_bin_smooth is not None):
k_mask_bin_smooth_ = "%6.4f"%self.bss_result.k_mask_bin_smooth[i_sel]
k_mask_bin_averaged_ = "%6.4f"%flex.mean(self.core.k_mask().select(sel))
d_ = self.d_spacings.data().select(sel)
d_min_ = flex.min(d_)
d_max_ = flex.max(d_)
n_ref_ = d_.size()
f_obs_ = self.f_obs.select(sel)
f_obs_mean_ = flex.mean(f_obs_.data())
k_isotropic_ = flex.mean(self.core.k_isotropic.select(sel))
k_anisotropic_ = flex.mean(self.core.k_anisotropic.select(sel))
cmpl_ = f_obs_.completeness(d_max=d_max_)*100.
r_ = bulk_solvent.r_factor(f_obs_.data(),f_model.select(sel),1)
print >> self.log, fmt%(ss_, d_max_, d_min_, cmpl_, n_ref_,
k_mask_bin_orig_, k_mask_bin_smooth_,k_mask_bin_averaged_,
k_isotropic_, k_anisotropic_, f_obs_mean_, r_)
def apply_back_trace_of_overall_exp_scale_matrix(self, xray_structure=None):
k,b=self.overall_isotropic_kb_estimate()
k_total = self.core.k_isotropic * self.core.k_anisotropic * \
self.core.k_isotropic_exp
k,b,r = mmtbx.bulk_solvent.fit_k_exp_b_to_k_total(k_total, self.ss, k, b)
if(r<0.7): self.k_exp_overall,self.b_exp_overall = k,b
if(xray_structure is None): return None
b_adj = 0
if([self.k_exp_overall,self.b_exp_overall].count(None)==0 and k != 0):
bs1 = xray_structure.extract_u_iso_or_u_equiv()*adptbx.u_as_b(1.)
def split(b_trace, xray_structure):
b_min = xray_structure.min_u_cart_eigenvalue()*adptbx.u_as_b(1.)
b_res = min(0, b_min + b_trace+1.e-6)
b_adj = b_trace-b_res
xray_structure.shift_us(b_shift = b_adj)
return b_adj, b_res
b_adj,b_res=split(b_trace=self.b_exp_overall,xray_structure=xray_structure)
k_new = self.k_exp_overall*flex.exp(-self.ss*b_adj)
bs2 = xray_structure.extract_u_iso_or_u_equiv()*adptbx.u_as_b(1.)
diff = bs2-bs1
assert approx_equal(flex.min(diff), flex.max(diff))
assert approx_equal(flex.max(diff), b_adj)
self.core = self.core.update(
k_isotropic = self.core.k_isotropic,
k_isotropic_exp = self.core.k_isotropic_exp/k_new,
k_masks = [m*flex.exp(-self.ss*b_adj) for m in self.core.k_masks])
return group_args(
xray_structure = xray_structure,
k_isotropic = self.k_isotropic(),
k_anisotropic = self.k_anisotropic(),
k_mask = self.k_masks(),
b_adj = b_adj)
def __init__(self, xray_structure, k_anisotropic, k_masks, ss):
self.xray_structure = xray_structure
self.k_anisotropic = k_anisotropic
self.k_masks = k_masks
self.ss = ss
#
k_total = self.k_anisotropic
r = scitbx.math.gaussian_fit_1d_analytical(x=flex.sqrt(self.ss), y=k_total)
k,b = r.a, r.b
#
k,b,r = mmtbx.bulk_solvent.fit_k_exp_b_to_k_total(k_total, self.ss, k, b)
k_exp_overall, b_exp_overall = None,None
if(r<0.7): k_exp_overall, b_exp_overall = k,b
if(self.xray_structure is None): return None
b_adj = 0
if([k_exp_overall, b_exp_overall].count(None)==0 and k != 0):
bs1 = self.xray_structure.extract_u_iso_or_u_equiv()*adptbx.u_as_b(1.)
def split(b_trace, xray_structure):
b_min = xray_structure.min_u_cart_eigenvalue()*adptbx.u_as_b(1.)
b_res = min(0, b_min + b_trace+1.e-6)
b_adj = b_trace-b_res
xray_structure.shift_us(b_shift = b_adj)
return b_adj, b_res
b_adj,b_res=split(b_trace=b_exp_overall,xray_structure=self.xray_structure)
k_new = k_exp_overall*flex.exp(-self.ss*b_adj)
bs2 = self.xray_structure.extract_u_iso_or_u_equiv()*adptbx.u_as_b(1.)
diff = bs2-bs1
assert approx_equal(flex.min(diff), flex.max(diff))
assert approx_equal(flex.max(diff), b_adj)
self.k_anisotropic = self.k_anisotropic/k_new
self.k_masks = [m*flex.exp(-self.ss*b_adj) for m in self.k_masks]
def get_summary (self) :
"""
Returns a simple object for harvesting statistics elsewhere.
"""
n_anom_peaks = None
if (self.anom_peaks is not None) :
n_anom_peaks = len(self.anom_peaks.heights)
n_water_peaks = n_water_anom_peaks = None
if (self.water_peaks is not None) :
n_water_peaks = len(self.water_peaks)
if (self.water_anom_peaks is not None) :
n_water_anom_peaks = len(self.water_anom_peaks)
hole_max = peak_max = None
if (len(self.peaks.heights) > 0) :
peak_max = flex.max(self.peaks.heights)
if (len(self.holes.heights) > 0) :
hole_max = flex.min(self.holes.heights)
n_non_water_anom_peaks = None
if (getattr(self, "non_water_anom_peaks", None) is not None) :
n_non_water_anom_peaks = len(self.non_water_anom_peaks)
return summary(
n_peaks_1=(self.peaks.heights > self.map_cutoff).count(True),
n_peaks_2=(self.peaks.heights > self.map_cutoff + 3).count(True),
n_peaks_3=(self.peaks.heights > self.map_cutoff + 6).count(True),
n_holes_1=(self.holes.heights < -self.map_cutoff).count(True),
n_holes_2=(self.holes.heights < -self.map_cutoff - 3).count(True),
n_holes_3=(self.holes.heights < -self.map_cutoff - 6).count(True),
peak_max=peak_max,
hole_max=hole_max,
n_anom_peaks=n_anom_peaks,
n_water_peaks=n_water_peaks,
n_water_anom_peaks=n_water_anom_peaks,
map_cutoff=self.map_cutoff,
anom_map_cutoff=self.anom_map_cutoff,
n_non_water_anom_peaks=n_non_water_anom_peaks)
def get_model_stat(file_name):
grm = restraints.get_grm(file_name = file_name)
r = model_statistics.geometry(
pdb_hierarchy = grm.pdb_hierarchy,
restraints_manager = grm.restraints_manager,
molprobity_scores = True)
# XXX very runtime inefficient
distances = flex.double()
xyz = grm.pdb_hierarchy.atoms().extract_xyz()
bond_proxies_simple = grm.restraints_manager.geometry.pair_proxies(
sites_cart = xyz).bond_proxies.simple
for i, site_i in enumerate(xyz):
for j, site_j in enumerate(xyz):
if(j>i):
bonded = False
for proxy in bond_proxies_simple:
p1 = list(proxy.i_seqs)
p2 = [i,j]
p1.sort()
p2.sort()
if(p1==p2): bonded=True
if(not bonded):
dist_ij = math.sqrt(
(site_i[0]-site_j[0])**2+
(site_i[1]-site_j[1])**2+
(site_i[2]-site_j[2])**2)
distances.append(dist_ij)
min_nonbonded_distance = flex.min(distances)
# bond(rmsd), bond(max), angle(rmsd), angle(max), etc..
#print r.b_mean, r.b_max, r.a_min, r.a_max, r.clashscore, min_nonbonded_distance
return min_nonbonded_distance, r.b_max , r.a_max
def get_mean_statistic_for_resolution (d_min, stat_type, range=0.2, out=None) :
if (out is None) :
out = sys.stdout
from scitbx.array_family import flex
pkl_file = libtbx.env.find_in_repositories(
relative_path = "chem_data/polygon_data/all_mvd.pickle",
test = os.path.isfile)
db = easy_pickle.load(pkl_file)
all_d_min = db['high_resolution']
stat_values = db[stat_type]
values_for_range = flex.double()
for (d_, v_) in zip(all_d_min, stat_values) :
try :
d = float(d_)
v = float(v_)
except ValueError : continue
else :
if (d > (d_min - range)) and (d < (d_min + range)) :
values_for_range.append(v)
h = flex.histogram(values_for_range, n_slots=10)
print >> out, " %s for d_min = %.3f - %.3f A" % (stat_names[stat_type], d_min-range,
d_min+range)
min = flex.min(values_for_range)
max = flex.max(values_for_range)
mean = flex.mean(values_for_range)
print >> out, " count: %d" % values_for_range.size()
print >> out, " min: %.2f" % min
print >> out, " max: %.2f" % max
print >> out, " mean: %.2f" % mean
print >> out, " histogram of values:"
h.show(prefix=" ")
return mean
def __init__(self,
evaluator,
population_size=50,
f=None,
cr=0.9,
eps=1e-2,
n_cross=1,
max_iter=10000,
monitor_cycle=200,
out=None,
show_progress=False,
show_progress_nth_cycle=1,
insert_solution_vector=None,
dither_constant=0.4):
self.dither=dither_constant
self.show_progress=show_progress
self.show_progress_nth_cycle=show_progress_nth_cycle
self.evaluator = evaluator
self.population_size = population_size
self.f = f
self.cr = cr
self.n_cross = n_cross
self.max_iter = max_iter
self.monitor_cycle = monitor_cycle
self.vector_length = evaluator.n
self.eps = eps
self.population = []
self.seeded = False
if insert_solution_vector is not None:
assert len( insert_solution_vector )==self.vector_length
self.seeded = insert_solution_vector
for ii in xrange(self.population_size):
self.population.append( flex.double(self.vector_length,0) )
self.scores = flex.double(self.population_size,1000)
self.optimize()
self.best_score = flex.min( self.scores )
self.best_vector = self.population[ flex.min_index( self.scores ) ]
self.evaluator.x = self.best_vector
if self.show_progress:
self.evaluator.print_status(
flex.min(self.scores),
flex.mean(self.scores),
self.population[ flex.min_index( self.scores ) ],
'Final')
def collect(O, rmsd_t_c, param_values): mins = [flex.min(a) for a in rmsd_t_c] means = [flex.mean(a) for a in rmsd_t_c] if (mins[0] < mins[1]): a = "t" else: a = "c" if (means[0] < means[1]): b = "t" else: b = "c" O.data[a+b].append(param_values)
def fsc_model_map(xray_structure, map, d_min, log=sys.stdout, radius=2.,
n_bins=30, prefix=""):
sgn = xray_structure.crystal_symmetry().space_group().type().number()
f_calc = xray_structure.structure_factors(d_min=d_min).f_calc()
def compute_mc(f_calc, map):
return f_calc.structure_factors_from_map(
map = map,
use_scale = True,
anomalous_flag = False,
use_sg = False)
sites_frac = xray_structure.sites_frac()
if(sgn==1):
mask = cctbx_maptbx_ext.mask(
sites_frac = sites_frac,
unit_cell = xray_structure.unit_cell(),
n_real = map.all(),
mask_value_inside_molecule = 1,
mask_value_outside_molecule = 0,
radii = flex.double(sites_frac.size(), radius))
mc = compute_mc(f_calc=f_calc, map=map)
if(sgn==1):
mc_masked = compute_mc(f_calc=f_calc, map=map*mask)
del mask
print >> log, prefix, "Overall (entire box): %7.4f"%\
f_calc.map_correlation(other = mc)
if(sgn==1):
cc = f_calc.map_correlation(other = mc_masked)
if(cc is not None): print >> log, prefix, "Around atoms (masked): %7.4f"%cc
dsd = f_calc.d_spacings().data()
if(dsd.size()>1500):
f_calc.setup_binner(n_bins = n_bins)
else:
f_calc.setup_binner(reflections_per_bin = dsd.size())
if(sgn==1):
print >> log, prefix, "Bin# Resolution (A) CC CC(masked)"
else:
print >> log, prefix, "Bin# Resolution (A) CC"
fmt1="%2d: %7.3f-%-7.3f %7.4f"
for i_bin in f_calc.binner().range_used():
sel = f_calc.binner().selection(i_bin)
d = dsd.select(sel)
d_min = flex.min(d)
d_max = flex.max(d)
n = d.size()
fc = f_calc.select(sel)
fo = mc.select(sel)
cc = fc.map_correlation(other = fo)
if(sgn==1):
fo_masked = mc_masked.select(sel)
cc_masked = fc.map_correlation(other = fo_masked)
if(cc_masked is not None and cc is not None):
fmt2="%2d: %7.3f-%-7.3f %7.4f %7.4f"
print >> log, prefix, fmt2%(i_bin, d_max, d_min, cc, cc_masked)
else:
fmt2="%2d: %7.3f-%-7.3f %s %s"
print >> log, prefix, fmt2%(i_bin, d_max, d_min, "none", "none")
else:
print >> log, prefix, fmt1%(i_bin, d_max, d_min, cc)
def optimize(self):
# initialise the population please
self.make_random_population()
# score the population please
self.score_population()
converged = False
monitor_score = flex.min( self.scores )
self.count = 0
while not converged:
self.evolve()
location = flex.min_index( self.scores )
if self.show_progress:
if self.count%self.show_progress_nth_cycle==0:
# make here a call to a custom print_status function in the evaluator function
# the function signature should be (min_target, mean_target, best vector)
self.evaluator.print_status(
flex.min(self.scores),
flex.mean(self.scores),
self.population[ flex.min_index( self.scores ) ],
self.count)
self.count += 1
if self.count%self.monitor_cycle==0:
if (monitor_score-flex.min(self.scores) ) < self.eps:
converged = True
else:
monitor_score = flex.min(self.scores)
rd = (flex.mean(self.scores) - flex.min(self.scores) )
rd = rd*rd/(flex.min(self.scores)*flex.min(self.scores) + self.eps )
if ( rd < self.eps ):
converged = True
if self.count>=self.max_iter:
converged =True
def process_file(file_object, n_slots, data_min, data_max, format_cutoffs):
data = flex.double()
for line in file_object.read().splitlines():
data.append(float(line))
print "total number of data points:", data.size()
if (data_min is None): data_min = flex.min(data)
if (data_max is None): data_max = flex.max(data)
flex.histogram(
data=data, n_slots=n_slots, data_min=data_min, data_max=data_max).show(
format_cutoffs=format_cutoffs)
def test_structure_generator():
p = pdb.input(source_info='string',lines=test_pdb)
sg = structure_generator()
sg.add_species(p,100)
sg.randomize()
t = sg.translations[0]
d = flex.double()
for i in xrange(len(t)):
for j in xrange(i+1,len(t)):
d.append( (flex.double(t[i]) - flex.double(t[j])).norm() )
assert ( flex.min(d) > (sg.min_separation + 2.0*sg.species[0].radius) )
def normalize_start_map(self):
rho = self.rho_obs.deep_copy()
if(self.start_map == "flat"):
rho = flex.double(flex.grid(self.n_real), 1./self.N)
elif(self.start_map == "lde"):
eps = flex.max(rho)/100.
selection_nonpositive = rho <= eps
rho = rho.set_selected(selection_nonpositive, eps)
elif(self.start_map == "min_shifted"):
rho = rho - flex.min(rho)
else: raise Sorry("Invalid initial map modification choice.")
return rho / flex.sum(rho)
def interpolate(self, x_array, y_array):
index_array = []
result_array = []
start_index_user = None
end_index_user = None
start_index_target = None
end_index_target = None
user_min = flex.min( x_array )
user_max = flex.max( x_array )
for jj,x in enumerate(self.target_x):
this_index = None
break_again = False
for index,this_x in enumerate(x_array):
if this_x - x >= 0:
if x >= user_min:
if x <= user_max:
this_index = index
if start_index_user is None:
start_index_user = this_index
if start_index_target is None:
start_index_target = jj
end_index_user = this_index
end_index_target = jj
break
index_array.append( this_index )
y = None
if this_index is not None:
if this_index == 0:
y = self.two_point_interpolate( x,
x_array[this_index ], y_array[this_index ],
x_array[this_index+1], y_array[this_index+1] )
elif this_index == len(x_array)-1:
y = self.two_point_interpolate( x,
x_array[this_index-1], y_array[this_index-1],
x_array[this_index], y_array[this_index] )
else:
y = self.parabolic_interpolate( x,
x_array[this_index-1], y_array[this_index-1],
x_array[this_index ], y_array[this_index ],
x_array[this_index+1], y_array[this_index+1] )
result_array.append( y )
n = len(result_array)
x = flex.double(self.target_x[start_index_target:end_index_target+1])
y = flex.double(result_array)
return x,y,(start_index_user,end_index_user),(start_index_target,end_index_target)
def run_simulation(
out,
six_dof_type,
r_is_qr,
mersenne_twister,
n_dynamics_steps,
delta_t):
sim = six_dof_simulation(
six_dof_type=six_dof_type,
r_is_qr=r_is_qr,
mersenne_twister=mersenne_twister)
sim_label = 'six_dof(type="%s", r_is_qr=%s)' % (
six_dof_type, str(sim.J.r_is_qr))
sim.check_d_pot_d_q()
sites_moved = [sim.sites_moved()]
e_pots = flex.double([sim.e_pot])
e_kins = flex.double([sim.e_kin])
for i_step in range(n_dynamics_steps):
sim.dynamics_step(delta_t=delta_t)
sites_moved.append(sim.sites_moved())
e_pots.append(sim.e_pot)
e_kins.append(sim.e_kin)
e_tots = e_pots + e_kins
sim.check_d_pot_d_q()
print(sim_label, file=out)
print("e_pot min, max:", min(e_pots), max(e_pots), file=out)
print("e_kin min, max:", min(e_kins), max(e_kins), file=out)
print("e_tot min, max:", min(e_tots), max(e_tots), file=out)
print("start e_tot:", e_tots[0], file=out)
print("final e_tot:", e_tots[-1], file=out)
ave = flex.sum(e_tots) / e_tots.size()
range_ = flex.max(e_tots) - flex.min(e_tots)
relative_range = range_ / ave
print("ave:", ave, file=out)
print("range:", range_, file=out)
print("relative range:", relative_range, file=out)
print(file=out)
out.flush()
if (out is sys.stdout):
l = sim_label \
.replace(' ', "") \
.replace('"', "") \
.replace("(", "_") \
.replace(")", "_") \
.replace(",", "_")
f = open("tmp_%02d_%02d_%s.xy" % (plot_prefix, plot_number[0], l), "w")
for es in [e_pots, e_kins, e_tots]:
for e in es:
print(e, file=f)
print("&", file=f)
f.close()
plot_number[0] += 1
return sim, sim_label, sites_moved, e_tots, relative_range
def __init__(self, rs_vectors, percentile=0.05):
from scitbx.array_family import flex
NEAR = 10
self.NNBIN = 5 # target number of neighbors per histogram bin
# nearest neighbor analysis
from annlib_ext import AnnAdaptor
query = flex.double()
for spot in rs_vectors: # spots, in reciprocal space xyz
query.append(spot[0])
query.append(spot[1])
query.append(spot[2])
assert len(
rs_vectors) > NEAR # Can't do nearest neighbor with too few spots
IS_adapt = AnnAdaptor(data=query, dim=3, k=1)
IS_adapt.query(query)
direct = flex.double()
for i in range(len(rs_vectors)):
direct.append(1.0 / math.sqrt(IS_adapt.distances[i]))
# determine the most probable nearest neighbor distance (direct space)
hst = flex.histogram(direct, n_slots=int(len(rs_vectors) / self.NNBIN))
centers = hst.slot_centers()
islot = hst.slots()
highest_bin_height = flex.max(islot)
most_probable_neighbor = centers[list(islot).index(highest_bin_height)]
if False: # to print out the histogramming analysis
smin, smax = flex.min(direct), flex.max(direct)
stats = flex.mean_and_variance(direct)
import sys
out = sys.stdout
print(" range: %6.2f - %.2f" % (smin, smax), file=out)
print(" mean: %6.2f +/- %6.2f on N = %d" %
(stats.mean(), stats.unweighted_sample_standard_deviation(),
direct.size()),
file=out)
hst.show(f=out, prefix=" ", format_cutoffs="%6.2f")
print("", file=out)
# determine the 5th-percentile direct-space distance
perm = flex.sort_permutation(direct, reverse=True)
percentile = direct[perm[int(percentile * len(rs_vectors))]]
MAXTOL = 1.5 # Margin of error for max unit cell estimate
self.max_cell = max(MAXTOL * most_probable_neighbor,
MAXTOL * percentile)
if False:
self.plot(direct)
def __init__(self, pdb_hierarchy, xray_structure, params, out=sys.stdout):
from cctbx import adptbx
from scitbx.array_family import flex
self.plot_range = params.plot_range
self.chains = []
self.residues = []
b_isos = xray_structure.extract_u_iso_or_u_equiv() * adptbx.u_as_b(1.0)
occ = pdb_hierarchy.atoms().extract_occ()
model = pdb_hierarchy.models()[0]
for chain in model.chains():
main_conf = chain.conformers()[0]
is_na = main_conf.is_na()
is_protein = main_conf.is_protein()
if (not is_protein) and (not is_na):
print >> out, "Skipping chain '%s' - not protein or DNA/RNA." % chain.id
continue
self.chains.append(chain.id)
self.residues.append([])
for residue_group in chain.residue_groups():
n_conformers = len(residue_group.atom_groups())
rg_i_seqs = residue_group.atoms().extract_i_seq()
rg_occ = residue_group.atoms().extract_occ()
if (params.average_b_over == "residue"):
use_i_seqs = rg_i_seqs
elif (params.average_b_over == "mainchain"):
use_i_seqs = []
if (is_protein):
for j_seq, atom in enumerate(residue_group.atoms()):
#alab = atom.fetch_labels()
if (atom.name in [" N ", " C ", " CA ", " O "]):
use_i_seqs.append(rg_i_seqs[j_seq])
else:
raise Sorry(
"Mainchain-only mode not supported for nucleic acids."
)
else:
use_i_seqs = []
if (is_protein):
for j_seq, atom in enumerate(residue_group.atoms()):
if (not atom.name
in [" N ", " C ", " CA ", " O "]):
use_i_seqs.append(rg_i_seqs[j_seq])
if (len(use_i_seqs) > 0):
has_partocc = ((flex.min(occ.select(use_i_seqs)) < 1.0)
and (n_conformers == 1))
res_info = residue_info(
chain_id=chain.id,
resseq=residue_group.resseq_as_int(),
icode=residue_group.icode,
has_altconf=(n_conformers > 1),
has_partocc=has_partocc,
avg_b=flex.mean(b_isos.select(use_i_seqs)))
self.residues[-1].append(res_info)
def exercise_eigensystem():
#random.seed(0)
for n in xrange(1, 10):
m = flex.double(flex.grid(n, n))
s = tntbx.eigensystem.real(m)
assert approx_equal(tuple(s.values()), [0] * n)
v = s.vectors()
for i in xrange(n):
for j in xrange(n):
x = 0
if (i == j): x = 1
#assert approx_equal(v[(i,j)], x)
v = []
for i in xrange(n):
j = (i * 13 + 17) % n
v.append(j)
m[i * (n + 1)] = j
s = tntbx.eigensystem.real(m)
if (n == 3):
ss = tntbx.eigensystem.real((m[0], m[4], m[8], m[1], m[2], m[5]))
assert approx_equal(s.values(), ss.values())
assert approx_equal(s.vectors(), ss.vectors())
v.sort()
v.reverse()
assert approx_equal(s.values(), v)
if (n > 1):
assert approx_equal(flex.min(s.vectors()), 0)
assert approx_equal(flex.max(s.vectors()), 1)
assert approx_equal(flex.sum(s.vectors()), n)
for t in xrange(10):
for i in xrange(n):
for j in xrange(i, n):
m[i * n + j] = random.random() - 0.5
if (i != j):
m[j * n + i] = m[i * n + j]
s = tntbx.eigensystem.real(m)
if (n == 3):
ss = tntbx.eigensystem.real(
(m[0], m[4], m[8], m[1], m[2], m[5]))
assert approx_equal(s.values(), ss.values())
assert approx_equal(s.vectors(), ss.vectors())
v = list(s.values())
v.sort()
v.reverse()
assert list(s.values()) == v
for i in xrange(n):
l = s.values()[i]
x = s.vectors()[i * n:i * n + n]
mx = matrix_mul(m, n, n, x, n, 1)
lx = [e * l for e in x]
assert approx_equal(mx, lx)
m = (1.4573362052597449, 1.7361052947659894, 2.8065584999742659,
-0.5387293498219814, -0.018204949672480729, 0.44956507395617257)
def run_simulation(
out,
six_dof_type,
r_is_qr,
mersenne_twister,
n_dynamics_steps,
delta_t):
sim = six_dof_simulation(
six_dof_type=six_dof_type,
r_is_qr=r_is_qr,
mersenne_twister=mersenne_twister)
sim_label = 'six_dof(type="%s", r_is_qr=%s)' % (
six_dof_type, str(sim.J.r_is_qr))
sim.check_d_pot_d_q()
sites_moved = [sim.sites_moved()]
e_pots = flex.double([sim.e_pot])
e_kins = flex.double([sim.e_kin])
for i_step in xrange(n_dynamics_steps):
sim.dynamics_step(delta_t=delta_t)
sites_moved.append(sim.sites_moved())
e_pots.append(sim.e_pot)
e_kins.append(sim.e_kin)
e_tots = e_pots + e_kins
sim.check_d_pot_d_q()
print >> out, sim_label
print >> out, "e_pot min, max:", min(e_pots), max(e_pots)
print >> out, "e_kin min, max:", min(e_kins), max(e_kins)
print >> out, "e_tot min, max:", min(e_tots), max(e_tots)
print >> out, "start e_tot:", e_tots[0]
print >> out, "final e_tot:", e_tots[-1]
ave = flex.sum(e_tots) / e_tots.size()
range = flex.max(e_tots) - flex.min(e_tots)
relative_range = range / ave
print >> out, "ave:", ave
print >> out, "range:", range
print >> out, "relative range:", relative_range
print >> out
out.flush()
if (out is sys.stdout):
l = sim_label \
.replace(' ', "") \
.replace('"', "") \
.replace("(", "_") \
.replace(")", "_") \
.replace(",", "_")
f = open("tmp_%02d_%02d_%s.xy" % (plot_prefix, plot_number[0], l), "w")
for es in [e_pots, e_kins, e_tots]:
for e in es:
print >> f, e
print >> f, "&"
f.close()
plot_number[0] += 1
return sim, sim_label, sites_moved, e_tots, relative_range
def get_binned_intensities(self, n_bins=100):
"""
Using self.ISIGI, bin the intensities using the following procedure:
1) Find the minimum and maximum intensity values.
2) Divide max-min by n_bins. This is the bin step size
The effect is
@param n_bins number of bins to use.
@return a tuple with an array of selections for each bin and an array of median
intensity values for each bin.
"""
print("Computing intensity bins.", end=' ', file=self.log)
ISIGI = self.scaler.ISIGI
meanI = ISIGI['mean_scaled_intensity']
sels = []
binned_intensities = []
if True:
# intensity range per bin is the same
min_meanI = flex.min(meanI)
step = (flex.max(meanI)-min_meanI)/n_bins
print("Bin size:", step, file=self.log)
self.bin_indices = flex.int(len(ISIGI), -1)
for i in range(n_bins):
if i+1 == n_bins:
sel = (meanI >= (min_meanI + step * i))
else:
sel = (meanI >= (min_meanI + step * i)) & (meanI < (min_meanI + step * (i+1)))
if sel.all_eq(False): continue
sels.append(sel)
self.bin_indices.set_selected(sel, len(sels)-1)
binned_intensities.append((step/2 + step*i)+min_meanI)
assert(self.bin_indices == -1).count(True) == False
else:
# n obs per bin is the same
sorted_meanI = meanI.select(flex.sort_permutation(meanI))
bin_size = len(meanI)/n_bins
for i in range(n_bins):
bin_min = sorted_meanI[int(i*bin_size)]
sel = meanI >= bin_min
if i+1 == n_bins:
bin_max = sorted_meanI[-1]
else:
bin_max = sorted_meanI[int((i+1)*bin_size)]
sel &= meanI < bin_max
sels.append(sel)
binned_intensities.append(bin_min + ((bin_max-bin_min)/2))
for i, (sel, intensity) in enumerate(zip(sels, binned_intensities)):
print("Bin %02d, number of observations: % 10d, midpoint intensity: %f"%(i, sel.count(True), intensity), file=self.log)
return sels, binned_intensities
def run(args=None):
dxtbx.util.encode_output_as_utf8()
datablocks = DataBlockFactory.from_args(args or sys.argv[1:])
assert len(datablocks) == 1
detectors = datablocks[0].unique_detectors()
assert len(detectors) == 1
detector = detectors[0]
assert len(detector) == 1
px_mm = detector[0].get_px_mm_strategy()
assert isinstance(px_mm, ParallaxCorrectedPxMmStrategy)
print("Mu: %f mm^-1 " % px_mm.mu())
print("t0: %f mm" % px_mm.t0())
image_size = detector[0].get_image_size()[::-1]
xcorr = flex.double(flex.grid(image_size))
ycorr = flex.double(flex.grid(image_size))
pixel_size = detector[0].get_pixel_size()
for j in range(xcorr.all()[0]):
for i in range(xcorr.all()[1]):
x1, y1 = detector[0].pixel_to_millimeter((i, j))
x0, y0 = i * pixel_size[0], j * pixel_size[1]
xcorr[j, i] = x1 - x0
ycorr[j, i] = y1 - y0
vmin = min([flex.min(xcorr), flex.min(ycorr)])
vmax = max([flex.max(xcorr), flex.max(ycorr)])
fig, ax = pylab.subplots()
pylab.subplot(121)
pylab.imshow(xcorr.as_numpy_array(),
interpolation="none",
vmin=vmin,
vmax=vmax)
pylab.subplot(122)
im = pylab.imshow(ycorr.as_numpy_array(),
interpolation="none",
vmin=vmin,
vmax=vmax)
fig.subplots_adjust(right=0.8)
cax = fig.add_axes([0.9, 0.1, 0.03, 0.8])
fig.colorbar(im, cax=cax)
pylab.show()
def exercise_eigensystem():
#random.seed(0)
for n in xrange(1,10):
m = flex.double(flex.grid(n,n))
s = tntbx.eigensystem.real(m)
assert approx_equal(tuple(s.values()), [0]*n)
v = s.vectors()
for i in xrange(n):
for j in xrange(n):
x = 0
if (i == j): x = 1
#assert approx_equal(v[(i,j)], x)
v = []
for i in xrange(n):
j = (i*13+17) % n
v.append(j)
m[i*(n+1)] = j
s = tntbx.eigensystem.real(m)
if (n == 3):
ss = tntbx.eigensystem.real((m[0],m[4],m[8],m[1],m[2],m[5]))
assert approx_equal(s.values(), ss.values())
assert approx_equal(s.vectors(), ss.vectors())
v.sort()
v.reverse()
assert approx_equal(s.values(), v)
if (n > 1):
assert approx_equal(flex.min(s.vectors()), 0)
assert approx_equal(flex.max(s.vectors()), 1)
assert approx_equal(flex.sum(s.vectors()), n)
for t in xrange(10):
for i in xrange(n):
for j in xrange(i,n):
m[i*n+j] = random.random() - 0.5
if (i != j):
m[j*n+i] = m[i*n+j]
s = tntbx.eigensystem.real(m)
if (n == 3):
ss = tntbx.eigensystem.real((m[0],m[4],m[8],m[1],m[2],m[5]))
assert approx_equal(s.values(), ss.values())
assert approx_equal(s.vectors(), ss.vectors())
v = list(s.values())
v.sort()
v.reverse()
assert list(s.values()) == v
for i in xrange(n):
l = s.values()[i]
x = s.vectors()[i*n:i*n+n]
mx = matrix_mul(m, n, n, x, n, 1)
lx = [e*l for e in x]
assert approx_equal(mx, lx)
m = (1.4573362052597449, 1.7361052947659894, 2.8065584999742659,
-0.5387293498219814, -0.018204949672480729, 0.44956507395617257)
def nsd(self,moving,d_moving=None):
if self.d_moving is None:
self.d_moving = self.get_mean_distance(moving)
if d_moving is not None:
self.d_moving = d_moving
# loop over all sites in fixed, find the minimum for each site
tot_rho_mf = 0
tot_rho_fm = 0
for site in moving:
dd = self.fixed-site
dd = flex.min( dd.norms() )
tot_rho_mf+=dd*dd
for site in self.fixed:
dd = moving-site
dd = flex.min( dd.norms() )
tot_rho_fm+=dd
tot_rho_fm = tot_rho_fm / (self.fixed.size()*self.d_fixed )
tot_rho_mf = tot_rho_mf / (moving.size()*self.d_moving )
result = smath.sqrt((tot_rho_fm+tot_rho_mf)/2.0)
return result
def weighted_means(self): min_score = flex.min( self.scores ) mw = flex.mean( self.scores ) self.scores = self.scores-min_score wghts = flex.exp( -self.scores*0.50 ) sw = 1e-12+flex.sum( wghts ) mrg = flex.sum(wghts*self.rg)/sw srg = flex.sum(wghts*self.rg*self.rg)/sw mi0 = flex.sum(wghts*self.i0)/sw si0 = flex.sum(wghts*self.i0*self.i0)/sw si0 = math.sqrt(si0-mi0*mi0) srg = math.sqrt(srg-mrg*mrg) return mrg,srg,mi0,si0,mw
def get_binned_intensities(self, n_bins=100):
"""
Using self.ISIGI, bin the intensities using the following procedure:
1) Find the minimum and maximum intensity values.
2) Divide max-min by n_bins. This is the bin step size
The effect is
@param n_bins number of bins to use.
@return a tuple with an array of selections for each bin and an array of median
intensity values for each bin.
"""
print("Computing intensity bins.", end=' ', file=self.log)
all_mean_Is = flex.double()
only_means = flex.double()
for hkl_id in range(self.scaler.n_refl):
hkl = self.scaler.miller_set.indices()[hkl_id]
if hkl not in self.scaler.ISIGI: continue
n = len(self.scaler.ISIGI[hkl])
# get scaled intensities
intensities = flex.double([self.scaler.ISIGI[hkl][i][0] for i in range(n)])
meanI = flex.mean(intensities)
only_means.append(meanI)
all_mean_Is.extend(flex.double([meanI]*n))
step = (flex.max(only_means)-flex.min(only_means))/n_bins
print("Bin size:", step, file=self.log)
sels = []
binned_intensities = []
min_all_mean_Is = flex.min(all_mean_Is)
for i in range(n_bins):
sel = (all_mean_Is > (min_all_mean_Is + step * i)) & (all_mean_Is < (min_all_mean_Is + step * (i+1)))
if sel.all_eq(False): continue
sels.append(sel)
binned_intensities.append((step/2 + step*i)+min(only_means))
for i, (sel, intensity) in enumerate(zip(sels, binned_intensities)):
print("Bin %02d, number of observations: % 10d, midpoint intensity: %f"%(i, sel.count(True), intensity), file=self.log)
return sels, binned_intensities
def check_adp(u_iso, step=10, out=None):
if (out is None):
out = sys.stdout
min_adp = flex.min(u_iso)
if (min_adp <= 0):
bad_i_seqs = []
for i_seq in range(len(u_iso)):
if (u_iso[i_seq] <= 0):
bad_i_seqs.append(i_seq)
return bad_i_seqs
i = 0
while i < u_iso.size():
if (i + step < u_iso.size()):
u_iso_i = u_iso[i:i + step]
else:
u_iso_i = u_iso[i:]
if (u_iso_i.size() >= step // 2):
min_adp = flex.min(u_iso)
max_adp = flex.max(u_iso)
if (abs(min_adp - max_adp) < 0.1):
raise Sorry("At least 10 bonded atoms have identical ADPs.")
i += step
return None
def run(self):
from dials.algorithms.image.fill_holes import simple_fill
from scitbx.array_family import flex
from math import sqrt
mask = flex.bool(flex.grid(100, 100), True)
data = flex.double(flex.grid(100, 100), True)
for j in range(100):
for i in range(100):
data[j, i] = 10 + j * 0.01 + i * 0.01
if sqrt((j - 50)**2 + (i - 50)**2) <= 10.5:
mask[j, i] = False
data[j, i] = 0
result = simple_fill(data, mask)
known = data.as_1d().select(mask.as_1d())
filled = result.as_1d().select(mask.as_1d() == False)
assert flex.max(filled) <= flex.max(known)
assert flex.min(filled) >= flex.min(known)
# Test passed
print 'OK'
def chebyshev_nodes(n, low=-1, high=1, include_limits=False):
x = flex.double(range(n)) + 1
x = (2.0 * x - 1.0) / n
x = x * math.pi / 2.0
x = -flex.cos(x)
if include_limits:
span = (flex.max(x) - flex.min(x)) / 2.0
x = x / span
x = 0.5 * (low + high) + 0.5 * (high - low) * x
if include_limits:
x[0] = low
x[n - 1] = high
return (x)
def main(filenames,
map_file,
npoints=192,
max_resolution=6,
reverse_phi=False):
rec_range = 1 / max_resolution
image = ImageFactory(filenames[0])
panel = image.get_detector()[0]
beam = image.get_beam()
s0 = beam.get_s0()
pixel_size = panel.get_pixel_size()
xlim, ylim = image.get_raw_data().all()
xy = recviewer.get_target_pixels(panel, s0, xlim, ylim, max_resolution)
s1 = panel.get_lab_coord(xy * pixel_size[0]) # FIXME: assumed square pixel
s1 = s1 / s1.norms() * (1 / beam.get_wavelength()) # / is not supported...
S = s1 - s0
grid = flex.double(flex.grid(npoints, npoints, npoints), 0)
cnts = flex.int(flex.grid(npoints, npoints, npoints), 0)
for filename in filenames:
print "Processing image", filename
try:
fill_voxels(ImageFactory(filename), grid, cnts, S, xy, reverse_phi,
rec_range)
except:
print " Failed to process. Skipped this."
recviewer.normalize_voxels(grid, cnts)
uc = uctbx.unit_cell((npoints, npoints, npoints, 90, 90, 90))
ccp4_map.write_ccp4_map(map_file, uc, sgtbx.space_group("P1"), (0, 0, 0),
grid.all(), grid,
flex.std_string(["cctbx.miller.fft_map"]))
return
from scitbx import fftpack
fft = fftpack.complex_to_complex_3d(grid.all())
grid_complex = flex.complex_double(reals=flex.pow2(grid),
imags=flex.double(grid.size(), 0))
grid_transformed = flex.abs(fft.backward(grid_complex))
print flex.max(grid_transformed), flex.min(
grid_transformed), grid_transformed.all()
ccp4_map.write_ccp4_map(map_file, uc, sgtbx.space_group("P1"), (0, 0, 0),
grid.all(), grid_transformed,
flex.std_string(["cctbx.miller.fft_map"]))
def set_file(self, file_name, hierarchy=None):
self.file_name = os.path.abspath(file_name)
from scitbx.array_family import flex
if (hierarchy is None):
from iotbx import file_reader
import iotbx.pdb
pdb_in = file_reader.any_file(
file_name,
force_type="pdb",
raise_sorry_if_errors=True,
raise_sorry_if_not_expected_format=True)
pdb_in.check_file_type("pdb")
hierarchy = pdb_in.file_object.hierarchy
if (len(hierarchy.models()) > 1):
raise Sorry("Multi-MODEL PDB files not supported.")
self._hierarchy = hierarchy
self.SetTitle("B-factors by chain for %s" % to_unicode(self.file_name))
self.file_txt.SetLabel(to_unicode(self.file_name))
chain_list = wx.ListCtrl(self.panel,
-1,
style=wx.LC_REPORT,
size=(480, 160))
chain_list.InsertColumn(0, "Chain info")
chain_list.InsertColumn(1, "Mean B-iso (range)")
chain_list.SetColumnWidth(0, 260)
chain_list.SetColumnWidth(1, 200)
for chain in hierarchy.models()[0].chains():
n_res = len(chain.residue_groups())
chain_atoms = chain.atoms()
n_atoms = len(chain_atoms)
main_conf = chain.conformers()[0]
chain_type = "other"
if (main_conf.is_protein()):
chain_type = "protein"
elif (main_conf.is_na()):
chain_type = "nucleic acid"
chain_info = "'%s' (%s, %d res., %d atoms)" % (
chain.id, chain_type, n_res, n_atoms)
b_iso = chain_atoms.extract_b()
b_max = flex.max(b_iso)
b_min = flex.min(b_iso)
b_mean = flex.mean(b_iso)
b_info = "%.2f (%.2f - %.2f)" % (b_mean, b_min, b_max)
item = chain_list.InsertStringItem(sys.maxunicode, chain_info)
chain_list.SetStringItem(item, 1, b_info)
self.panel_sizer.Add(chain_list, 1, wx.EXPAND | wx.ALL, 5)
self.panel.Layout()
self.panel_sizer.Fit(self.panel)
self.Fit()
def exercise_reference_impl_long(n_dynamics_steps, out):
sim = fmri.simulation()
e_tots = flex.double([sim.e_tot])
print >> out, "i_step, [e_pot, e_kin_ang, e_kin_lin, e_kin, e_tot]"
def show(i_step):
print >> out, \
i_step, [sim.e_pot, sim.e_kin_ang, sim.e_kin_lin, sim.e_kin, sim.e_tot]
out.flush()
n_show = max(1, n_dynamics_steps // 10)
for i_step in xrange(n_dynamics_steps):
sim.dynamics_step(delta_t=0.001)
e_tots.append(sim.e_tot)
if (i_step % n_show == 0):
show(i_step)
show(n_dynamics_steps)
print >> out
print >> out, "number of dynamics steps:", n_dynamics_steps
print >> out, "e_tot start:", e_tots[0]
print >> out, " final:", e_tots[-1]
print >> out, " min:", flex.min(e_tots)
print >> out, " max:", flex.max(e_tots)
print >> out, " max-min:", flex.max(e_tots) - flex.min(e_tots)
print >> out
out.flush()
def exercise_mask_data_1(space_group_info, n_sites=100):
from cctbx import maptbx
from cctbx.masks import vdw_radii_from_xray_structure
for d_min in [1, 1.5, 2.1]:
for resolution_factor in [1. / 2, 1. / 3, 1. / 4, 1. / 5]:
xrs = random_structure.xray_structure(
space_group_info=space_group_info,
elements=(("O", "N", "C") * (n_sites // 3 + 1))[:n_sites],
volume_per_atom=30,
min_distance=1)
atom_radii = vdw_radii_from_xray_structure(xray_structure=xrs)
asu_mask = masks.atom_mask(unit_cell=xrs.unit_cell(),
group=xrs.space_group(),
resolution=d_min,
grid_step_factor=resolution_factor,
solvent_radius=1.0,
shrink_truncation_radius=1.0)
asu_mask.compute(xrs.sites_frac(), atom_radii)
mask_data = asu_mask.mask_data_whole_uc()
assert flex.min(mask_data) == 0.0
# It's not just 0 and 1 ...
assert flex.max(mask_data) == xrs.space_group().order_z()
# In fact, it is a mixture ...
if 0: # XXX this will rightfully crash
mask_data_ = mask_data / xrs.space_group().order_z()
s0 = mask_data_ < 0.5
s1 = mask_data_ > 0.5
if (mask_data_.size() != s0.count(True) + s1.count(True)):
for d in mask_data_:
if (d != 0 and d != 1):
print(d, xrs.space_group().order_z())
assert mask_data_.size(
) == s0.count(True) + s1.count(True), [
mask_data_.size() - (s0.count(True) + s1.count(True))
]
if (
0
): # XXX This would crash with the message: "... The grid is not ..."
cr_gr = maptbx.crystal_gridding(
unit_cell=xrs.unit_cell(),
d_min=d_min,
resolution_factor=resolution_factor)
asu_mask = masks.atom_mask(unit_cell=xrs.unit_cell(),
space_group=xrs.space_group(),
gridding_n_real=cr_gr.n_real(),
solvent_radius=1.0,
shrink_truncation_radius=1.0)
asu_mask.compute(xrs.sites_frac(), atom_radii)
def write_image(file_name='image.png',
detector_size=None,
image_data=None,
max_value=2.0**8 - 1):
assert ((detector_size[0] * detector_size[1]) == len(image_data))
working_image_data = image_data.deep_copy()
min_image_value = flex.min(working_image_data)
if (min_image_value > 0.0):
working_image_data = flex.fabs(working_image_data / min_image_value -
1.0)
max_image_value = flex.max(working_image_data)
working_image_data = max_value / max_image_value * working_image_data
image = Image.new('L', detector_size)
image.putdata(working_image_data)
image = ImageOps.invert(image)
image.save(file_name)
def find_candidate_basis_vectors(self):
self.d_min = self.params.refinement_protocol.d_min_start
sel = self.reflections["id"] == -1
if self.d_min is not None:
sel &= 1 / self.reflections["rlp"].norms() > self.d_min
reflections = self.reflections.select(sel)
self.candidate_basis_vectors, used_in_indexing = self._basis_vector_search_strategy.find_basis_vectors(
reflections["rlp"])
self._used_in_indexing = sel.iselection().select(used_in_indexing)
if self.d_min is None:
self.d_min = flex.min(
1 /
self.reflections["rlp"].select(self._used_in_indexing).norms())
self.debug_show_candidate_basis_vectors()
return self.candidate_basis_vectors
def fit_data(p):
fls = fourier_legendre_series()
fls.read_polynomials(p.fls_data)
cos_sq = p.q*p.wavelength/(4.0*math.pi)
cos_sq = cos_sq*cos_sq
sin_sq = 1.0 - cos_sq
fit_x = cos_sq + sin_sq*flex.cos(p.x)
fit_ac = p.ac.deep_copy()
fit_x, fit_ac = zip(*sorted(zip(fit_x,fit_ac)))
fit_x = flex.double(fit_x)
fit_ac = flex.double(fit_ac)
fit_c = fls.compute_coefficients(p.fit_order,fit_ac,fit_x)
fit_c = set_odd_coefficients_to_zero(fit_c)
fit_v = fls.compute_coefficient_variances(p.fit_order,p.v,fit_x)
fit_v = set_odd_coefficients_to_zero(fit_v)
if (p.minimize):
nz_c = flex.double()
for k in xrange(0,len(fit_c),2):
if (fit_c[k] < 0.0):
fit_c[k] = -fit_c[k]
nz_c.append(fit_c[k])
m = lbfgs_optimizer(fit_ac,flex.sqrt(p.v),nz_c,fit_x,fls)
nz_v = m.estimate_asymptotic_variance(m.x)
assert(nz_c.size() == nz_v.size())
count = 0
for k in xrange(0,len(fit_c),2):
fit_c[k] = m.x[count]
fit_v[k] = nz_v[count]
count += 1
f_min = 1.0
f_max = 1.0
if (p.standardize):
# standardize fitted curve to have a min of 1.0 and max of 2.0
old_f = fls.compute_function(fit_c,fit_x)
# assume f is positive
f_min = flex.min(old_f)
f_max = flex.max(flex.fabs(old_f / f_min - 1.0))
scales = (f_min,f_max)
return fit_c, fit_v, scales
def is_within(dist, coords_1, coords_2):
"""Checks if any of coords_1 is within dist of coords_2"""
dist_sq = dist**2
if len(coords_2) < len(coords_1):
i1 = coords_2
i2 = coords_1
else:
i1 = coords_1
i2 = coords_2
for c1 in i1:
diffs = i2 - c1
min_d_sq = flex.min(diffs.dot())
if min_d_sq < dist_sq:
return True
return False
def __init__(self,
rawdata,
projection_vector,
spotfinder_spot,
verbose=False):
# projection vector is either the radial or azimuthal unit vector
# at a specific Bragg spot position
model_center = col(
(spotfinder_spot.ctr_mass_x(), spotfinder_spot.ctr_mass_y()))
px_x, px_y = project_2d_response_onto_line(projection_vector)
point_projections = flex.double()
pixel_values = flex.double()
for point in spotfinder_spot.bodypixels:
point_projection = (col(
(point.x, point.y)) - model_center).dot(projection_vector)
point_projections.append(point_projection)
pxval = rawdata[(point.x, point.y)]
if verbose:
print "point_projection", point_projection,
print "signal", pxval
pixel_values.append(pxval)
Lmin = flex.min(point_projections)
Lmax = flex.max(point_projections)
#print "Range %6.2f"%(Lmax-Lmin)
Rmin = round(Lmin - 2.0, 1)
Rmax = round(Lmax + 2.0, 1)
#print "Range %6.2f"%(Rmax-Rmin)
def histogram_bin(j):
return int(10. * (j - Rmin)) # bin units of 1/10 pixel
histo_x = flex.double((int(10 * (Rmax - Rmin))))
histo_y = flex.double(len(histo_x))
for ihis in xrange(len(histo_x)):
histo_x[ihis] = Rmin + 0.1 * ihis
for ipp, point_projection in enumerate(point_projections):
value = pixel_values[ipp]
for isample in xrange(len(px_x)):
histo_y[int(10 * (point_projection + px_x[isample] -
Rmin))] += value * px_y[isample]
self.histo_x = histo_x
self.histo_y = histo_y
def show_residual(self, plot=False):
F = self.sb.data.focus()
if plot:
# Residual
for x in range(F[1]):
for y in range(F[2]):
background = self.a[0] * x + self.a[1] * y + self.a[2]
model = background + self.a[3] * self.roi[x, y]
print("%4.0f" % (self.sb.data[0, x, y] - model), end=' ')
print()
print()
# analyse the variance
print(list(self.a))
diff = flex.double()
approx_poisson_sigma = flex.double()
for x in range(F[1]):
for y in range(F[2]):
background = self.a[0] * x + self.a[1] * y + self.a[2]
model = background + self.a[3] * self.roi[x, y]
diffdata = self.sb.data[0, x, y] - model
diff.append(diffdata)
if model < 0.:
model = 0. # complete kludge avoid a math domain error
approx_poisson_sigma.append(math.sqrt(model))
MV = flex.mean_and_variance(diff)
fmin, fmax = flex.min(diff), flex.max(diff)
print("residual Min=%4.0f, Mean=%4.0f, Max=%4.0f" %
(fmin, MV.mean(), fmax))
print("residual stddev=%4.0f" %
(MV.unweighted_sample_standard_deviation()))
print("ML mean Poisson stddev=%4.0f" %
(flex.mean(approx_poisson_sigma)))
if plot:
from matplotlib import pyplot as plt
n, bins, patches = plt.hist(diff,
40,
normed=1,
facecolor='g',
alpha=0.75)
plt.xlabel('Model vs. Observation Residual')
plt.ylabel('Probability')
plt.title('Histogram of Model Residual')
plt.show()
def plot_rij_histogram(rij_matrix, key="cosym_rij_histogram"):
"""Plot a histogram of the rij values.
Args:
plot_name (str): The file name to save the plot to.
If this is not defined then the plot is displayed in interactive mode.
"""
rij = rij_matrix.as_1d()
rij = rij.select(rij != 0)
hist = flex.histogram(
rij,
data_min=min(-1, flex.min(rij)),
data_max=max(1, flex.max(rij)),
n_slots=100,
)
d = {
key: {
"data": [{
"x": list(hist.slot_centers()),
"y": list(hist.slots()),
"type": "bar",
"name": "Rij histogram",
}],
"layout": {
"title": "Distribution of values in the Rij matrix",
"xaxis": {
"title": "r<sub>ij</sub>"
},
"yaxis": {
"title": "Frequency"
},
"bargap": 0,
},
"help":
"""\
A histogram of the values of the Rij matrix of pairwise correlation coefficients. A
unimodal distribution of values may suggest that no indexing ambiguity is evident,
whereas a bimodal distribution can be indicative of the presence of an indexing
ambiguity.
""",
}
}
return d
def output_image(flex_img, filename, invert=False, scale=False):
import Image
flex_img = flex_img.deep_copy()
flex_img -= flex.min(flex_img)
if scale:
img_max_value = 2**16
scale = img_max_value / flex.max(flex_img)
flex_img = flex_img.as_double() * scale
flex_img = flex_img
if invert:
img_max_value = 2**16
flex_img = img_max_value - flex_img # invert image for display
dim = flex_img.all()
#easy_pickle.dump("%s/avg_img.pickle" %output_dirname, flex_img)
byte_str = flex_img.slice_to_byte_str(0, flex_img.size())
im = Image.fromstring(mode="I", size=(dim[1], dim[0]), data=byte_str)
im = im.crop((0, 185, 391, 370))
#im.save("avg.tiff", "TIFF") # XXX This does not work (phenix.python -Qnew option)
im.save(filename, "PNG")
def __init__(self, xray_structure, k_anisotropic, k_masks, ss):
self.xray_structure = xray_structure
self.k_anisotropic = k_anisotropic
self.k_masks = k_masks
self.ss = ss
#
k_total = self.k_anisotropic
r = scitbx.math.gaussian_fit_1d_analytical(x=flex.sqrt(self.ss),
y=k_total)
k, b = r.a, r.b
#
k, b, r = mmtbx.bulk_solvent.fit_k_exp_b_to_k_total(
k_total, self.ss, k, b)
k_exp_overall, b_exp_overall = None, None
if (r < 0.7): k_exp_overall, b_exp_overall = k, b
if (self.xray_structure is None): return None
b_adj = 0
if ([k_exp_overall, b_exp_overall].count(None) == 0 and k != 0):
bs1 = self.xray_structure.extract_u_iso_or_u_equiv(
) * adptbx.u_as_b(1.)
def split(b_trace, xray_structure):
b_min = xray_structure.min_u_cart_eigenvalue() * adptbx.u_as_b(
1.)
b_res = min(0, b_min + b_trace + 1.e-6)
b_adj = b_trace - b_res
xray_structure.shift_us(b_shift=b_adj)
return b_adj, b_res
b_adj, b_res = split(b_trace=b_exp_overall,
xray_structure=self.xray_structure)
k_new = k_exp_overall * flex.exp(-self.ss * b_adj)
bs2 = self.xray_structure.extract_u_iso_or_u_equiv(
) * adptbx.u_as_b(1.)
diff = bs2 - bs1
assert approx_equal(flex.min(diff), flex.max(diff))
assert approx_equal(flex.max(diff), b_adj)
self.k_anisotropic = self.k_anisotropic / k_new
self.k_masks = [
m * flex.exp(-self.ss * b_adj) for m in self.k_masks
]
def show(self):
b = self.bss_result
print(" Statistics in resolution bins:", file=self.log)
#assert k_mask.size() == len(self.bin_selections)
fmt = " %7.5f %6.2f -%6.2f %5.1f %5d %-6s %-6s %-6s %6.3f %6.3f %8.2f %6.4f"
f_model = self.core.f_model.data()
print(
" s^2 Resolution Compl Nrefl k_mask k_iso k_ani <Fobs> R",
file=self.log)
print(" (A) (%) orig smooth average",
file=self.log)
k_mask_bin_orig_ = str(None)
k_mask_bin_smooth_ = str(None)
k_mask_bin_approx_ = str(None)
for i_sel, cas in enumerate(self.cores_and_selections):
selection, core, selection_use, sel_work = cas
sel = sel_work
ss_ = self.ss_bin_values[i_sel][2]
if (b is not None and self.bss_result.k_mask_bin_orig is not None):
k_mask_bin_orig_ = "%6.4f" % self.bss_result.k_mask_bin_orig[
i_sel]
if (b is not None
and self.bss_result.k_mask_bin_smooth is not None):
k_mask_bin_smooth_ = "%6.4f" % self.bss_result.k_mask_bin_smooth[
i_sel]
k_mask_bin_averaged_ = "%6.4f" % flex.mean(
self.core.k_mask().select(sel))
d_ = self.d_spacings.data().select(sel)
d_min_ = flex.min(d_)
d_max_ = flex.max(d_)
n_ref_ = d_.size()
f_obs_ = self.f_obs.select(sel)
f_obs_mean_ = flex.mean(f_obs_.data())
k_isotropic_ = flex.mean(self.core.k_isotropic.select(sel))
k_anisotropic_ = flex.mean(self.core.k_anisotropic.select(sel))
cmpl_ = f_obs_.completeness(d_max=d_max_) * 100.
r_ = bulk_solvent.r_factor(f_obs_.data(), f_model.select(sel), 1)
print(fmt % (ss_, d_max_, d_min_, cmpl_, n_ref_, k_mask_bin_orig_,
k_mask_bin_smooth_, k_mask_bin_averaged_,
k_isotropic_, k_anisotropic_, f_obs_mean_, r_),
file=self.log)
def do_sigma_scaling(self):
# Divide each pixel value by it's dark standard deviation. Since we are led
# to believe that the standard deviation of a pixel is proportional to the
# gain of said pixel, this approximates a gain correction.
assert self.dark_img is not None
assert self.gain_map is None # not appropriate to do sigma scaling and gain correction at the same time!
flex_cspad_img = self.cspad_img.as_double()
flex_cspad_img_sel = flex_cspad_img.as_1d().select(self.dark_mask.as_1d())
flex_dark_stddev = self.dark_stddev.select(self.dark_mask.as_1d()).as_double()
assert flex_dark_stddev.count(0) == 0
flex_dark_stddev /= flex.mean(flex_dark_stddev)
flex_cspad_img_sel /= flex_dark_stddev
flex_cspad_img.as_1d().set_selected(self.dark_mask.as_1d().iselection(), flex_cspad_img_sel)
self.cspad_img = flex_cspad_img
if 0: # for debugging
from matplotlib import pyplot
hist_min, hist_max = flex.min(flex_cspad_img_sel.as_double()), flex.max(flex_cspad_img_sel.as_double())
print hist_min, hist_max
n_slots = 100
n, bins, patches = pyplot.hist(flex_cspad_img_sel.as_1d().as_numpy_array(), bins=n_slots, range=(hist_min, hist_max))
pyplot.show()
def _batch_bins_and_data(batches, values, function_to_apply):
"""Apply function to the data from each batch.
Return the list of the batch bins and the value for each bin.
"""
batch_bins = []
data = []
i_batch_start = 0
current_batch = flex.min(batches)
n_ref = batches.size()
for i_ref in range(n_ref + 1):
if i_ref == n_ref or batches[i_ref] != current_batch:
assert batches[i_batch_start:i_ref].all_eq(current_batch)
values_for_batch = values[i_batch_start:i_ref]
data.append(function_to_apply(values_for_batch))
batch_bins.append(current_batch)
i_batch_start = i_ref
if i_ref < n_ref:
current_batch = batches[i_batch_start]
return batch_bins, data
