def __init__ (self, pdb_hierarchy, xray_structure, fmodel, distance_cutoff=4.0, collect_all=True, molprobity_map_params=None) : validation.__init__(self) from mmtbx.real_space_correlation import extract_map_stats_for_single_atoms from cctbx import adptbx from scitbx.matrix import col self.n_bad = 0 self.n_heavy = 0 pdb_atoms = pdb_hierarchy.atoms() if(len(pdb_atoms)>1): assert (not pdb_atoms.extract_i_seq().all_eq(0)) unit_cell = xray_structure.unit_cell() pair_asu_table = xray_structure.pair_asu_table( distance_cutoff = distance_cutoff) asu_mappings = pair_asu_table.asu_mappings() asu_table = pair_asu_table.table() u_isos = xray_structure.extract_u_iso_or_u_equiv() occupancies = xray_structure.scatterers().extract_occupancies() sites_cart = xray_structure.sites_cart() sites_frac = xray_structure.sites_frac() sel_cache = pdb_hierarchy.atom_selection_cache() water_sel = sel_cache.selection("resname HOH and name O") if (molprobity_map_params is not None): # assume parameters have been validated (symmetry of pdb and map matches) two_fofc_map = None fc_map = None d_min = None crystal_gridding = None # read two_fofc_map if (molprobity_map_params.map_file_name is not None): f = any_file(molprobity_map_params.map_file_name) two_fofc_map = f.file_object.map_data() d_min = molprobity_map_params.d_min crystal_gridding = maptbx.crystal_gridding( f.file_object.unit_cell(), space_group_info=space_group_info(f.file_object.space_group_number), pre_determined_n_real=f.file_object.unit_cell_grid) elif (molprobity_map_params.map_coefficients_file_name is not None): f = any_file(molprobity_map_params.map_coefficients_file_name) fourier_coefficients = f.file_server.get_miller_array( molprobity_map_params.map_coefficients_label) crystal_symmetry = fourier_coefficients.crystal_symmetry() d_min = fourier_coefficients.d_min() crystal_gridding = maptbx.crystal_gridding( crystal_symmetry.unit_cell(), d_min, resolution_factor=0.25, space_group_info=crystal_symmetry.space_group_info()) two_fofc_map = miller.fft_map( crystal_gridding=crystal_gridding, fourier_coefficients=fourier_coefficients).apply_sigma_scaling().\ real_map_unpadded() # calculate fc_map assert( (d_min is not None) and (crystal_gridding is not None) ) f_calc = xray_structure.structure_factors(d_min=d_min).f_calc() fc_map = miller.fft_map(crystal_gridding=crystal_gridding, fourier_coefficients=f_calc) fc_map = fc_map.apply_sigma_scaling().real_map_unpadded() map_stats = extract_map_stats_for_single_atoms( pdb_atoms=pdb_atoms, xray_structure=xray_structure, fmodel=None, selection=water_sel, fc_map=fc_map, two_fofc_map=two_fofc_map) else: map_stats = extract_map_stats_for_single_atoms( pdb_atoms=pdb_atoms, xray_structure=xray_structure, fmodel=fmodel, selection=water_sel) waters = [] for i_seq, atom in enumerate(pdb_atoms) : if (water_sel[i_seq]) : rt_mx_i_inv = asu_mappings.get_rt_mx(i_seq, 0).inverse() self.n_total += 1 asu_dict = asu_table[i_seq] nearest_atom = nearest_contact = None for j_seq, j_sym_groups in asu_dict.items() : atom_j = pdb_atoms[j_seq] site_j = sites_frac[j_seq] # Filter out hydrogens if atom_j.element.upper().strip() in ["H", "D"]: continue for j_sym_group in j_sym_groups: rt_mx = rt_mx_i_inv.multiply(asu_mappings.get_rt_mx(j_seq, j_sym_group[0])) site_ji = rt_mx * site_j site_ji_cart = xray_structure.unit_cell().orthogonalize(site_ji) vec_i = col(atom.xyz) vec_ji = col(site_ji_cart) dxyz = abs(vec_i - vec_ji) if (nearest_contact is None) or (dxyz < nearest_contact) : nearest_contact = dxyz nearest_atom = atom_info(pdb_atom=atom_j, symop=rt_mx) w = water( pdb_atom=atom, b_iso=adptbx.u_as_b(u_isos[i_seq]), occupancy=occupancies[i_seq], nearest_contact=nearest_contact, nearest_atom=nearest_atom, score=map_stats.two_fofc_ccs[i_seq], fmodel=map_stats.fmodel_values[i_seq], two_fofc=map_stats.two_fofc_values[i_seq], fofc=map_stats.fofc_values[i_seq], anom=map_stats.anom_values[i_seq], n_hbonds=None) # TODO if (w.is_bad_water()) : w.outlier = True self.n_bad += 1 elif (w.is_heavy_atom()) : w.outlier = True self.n_heavy += 1 if (w.outlier) or (collect_all) : self.results.append(w) self.n_outliers = len(self.results)
def __init__(self, pdb_hierarchy, xray_structure, fmodel, distance_cutoff=4.0, collect_all=True, molprobity_map_params=None): validation.__init__(self) from mmtbx.real_space_correlation import extract_map_stats_for_single_atoms from cctbx import adptbx from scitbx.matrix import col self.n_bad = 0 self.n_heavy = 0 pdb_atoms = pdb_hierarchy.atoms() if (len(pdb_atoms) > 1): assert (not pdb_atoms.extract_i_seq().all_eq(0)) unit_cell = xray_structure.unit_cell() pair_asu_table = xray_structure.pair_asu_table( distance_cutoff=distance_cutoff) asu_mappings = pair_asu_table.asu_mappings() asu_table = pair_asu_table.table() u_isos = xray_structure.extract_u_iso_or_u_equiv() occupancies = xray_structure.scatterers().extract_occupancies() sites_frac = xray_structure.sites_frac() sel_cache = pdb_hierarchy.atom_selection_cache() water_sel = sel_cache.selection("water") if (molprobity_map_params is not None): # assume parameters have been validated (symmetry of pdb and map matches) two_fofc_map = None fc_map = None d_min = None crystal_gridding = None # read two_fofc_map if (molprobity_map_params.map_file_name is not None): f = any_file(molprobity_map_params.map_file_name) two_fofc_map = f.file_object.map_data() d_min = molprobity_map_params.d_min crystal_gridding = maptbx.crystal_gridding( f.file_object.unit_cell(), space_group_info=space_group_info( f.file_object.space_group_number), pre_determined_n_real=f.file_object.unit_cell_grid) pdb_atoms = pdb_hierarchy.atoms() xray_structure = pdb_hierarchy.extract_xray_structure( crystal_symmetry=f.crystal_symmetry()) unit_cell = xray_structure.unit_cell() # check for origin shift # --------------------------------------------------------------------- soin = maptbx.shift_origin_if_needed( map_data=two_fofc_map, sites_cart=xray_structure.sites_cart(), crystal_symmetry=xray_structure.crystal_symmetry()) two_fofc_map = soin.map_data xray_structure.set_sites_cart(soin.sites_cart) # --------------------------------------------------------------------- pair_asu_table = xray_structure.pair_asu_table( distance_cutoff=distance_cutoff) asu_mappings = pair_asu_table.asu_mappings() asu_table = pair_asu_table.table() u_isos = xray_structure.extract_u_iso_or_u_equiv() occupancies = xray_structure.scatterers().extract_occupancies() sites_frac = xray_structure.sites_frac() sel_cache = pdb_hierarchy.atom_selection_cache() water_sel = sel_cache.selection("water") elif (molprobity_map_params.map_coefficients_file_name is not None): f = any_file(molprobity_map_params.map_coefficients_file_name) fourier_coefficients = f.file_server.get_miller_array( molprobity_map_params.map_coefficients_label) crystal_symmetry = fourier_coefficients.crystal_symmetry() d_min = fourier_coefficients.d_min() crystal_gridding = maptbx.crystal_gridding( crystal_symmetry.unit_cell(), d_min, resolution_factor=0.25, space_group_info=crystal_symmetry.space_group_info()) two_fofc_map = miller.fft_map( crystal_gridding=crystal_gridding, fourier_coefficients=fourier_coefficients).apply_sigma_scaling().\ real_map_unpadded() # calculate fc_map assert ((d_min is not None) and (crystal_gridding is not None)) f_calc = xray_structure.structure_factors(d_min=d_min).f_calc() fc_map = miller.fft_map(crystal_gridding=crystal_gridding, fourier_coefficients=f_calc) fc_map = fc_map.apply_sigma_scaling().real_map_unpadded() map_stats = extract_map_stats_for_single_atoms( pdb_atoms=pdb_atoms, xray_structure=xray_structure, fmodel=None, selection=water_sel, fc_map=fc_map, two_fofc_map=two_fofc_map) else: map_stats = extract_map_stats_for_single_atoms( pdb_atoms=pdb_atoms, xray_structure=xray_structure, fmodel=fmodel, selection=water_sel) waters = [] for i_seq, atom in enumerate(pdb_atoms): if (water_sel[i_seq]): rt_mx_i_inv = asu_mappings.get_rt_mx(i_seq, 0).inverse() self.n_total += 1 asu_dict = asu_table[i_seq] nearest_atom = nearest_contact = None for j_seq, j_sym_groups in asu_dict.items(): atom_j = pdb_atoms[j_seq] site_j = sites_frac[j_seq] # Filter out hydrogens if atom_j.element.upper().strip() in ["H", "D"]: continue for j_sym_group in j_sym_groups: rt_mx = rt_mx_i_inv.multiply( asu_mappings.get_rt_mx(j_seq, j_sym_group[0])) site_ji = rt_mx * site_j site_ji_cart = xray_structure.unit_cell( ).orthogonalize(site_ji) vec_i = col(atom.xyz) vec_ji = col(site_ji_cart) dxyz = abs(vec_i - vec_ji) if (nearest_contact is None) or (dxyz < nearest_contact): nearest_contact = dxyz nearest_atom = atom_info(pdb_atom=atom_j, symop=rt_mx) w = water(pdb_atom=atom, b_iso=adptbx.u_as_b(u_isos[i_seq]), occupancy=occupancies[i_seq], nearest_contact=nearest_contact, nearest_atom=nearest_atom, score=map_stats.two_fofc_ccs[i_seq], fmodel=map_stats.fmodel_values[i_seq], two_fofc=map_stats.two_fofc_values[i_seq], fofc=map_stats.fofc_values[i_seq], anom=map_stats.anom_values[i_seq], n_hbonds=None) # TODO if (w.is_bad_water()): w.outlier = True self.n_bad += 1 elif (w.is_heavy_atom()): w.outlier = True self.n_heavy += 1 if (w.outlier) or (collect_all): self.results.append(w) self.n_outliers = len(self.results)
def __init__ (self, pdb_hierarchy, xray_structure, fmodel, distance_cutoff=4.0, collect_all=True) : validation.__init__(self) from mmtbx.real_space_correlation import extract_map_stats_for_single_atoms from cctbx import adptbx from scitbx.matrix import col self.n_bad = 0 self.n_heavy = 0 pdb_atoms = pdb_hierarchy.atoms() if(len(pdb_atoms)>1): assert (not pdb_atoms.extract_i_seq().all_eq(0)) unit_cell = xray_structure.unit_cell() pair_asu_table = xray_structure.pair_asu_table( distance_cutoff = distance_cutoff) asu_mappings = pair_asu_table.asu_mappings() asu_table = pair_asu_table.table() u_isos = xray_structure.extract_u_iso_or_u_equiv() occupancies = xray_structure.scatterers().extract_occupancies() sites_cart = xray_structure.sites_cart() sites_frac = xray_structure.sites_frac() sel_cache = pdb_hierarchy.atom_selection_cache() water_sel = sel_cache.selection("resname HOH and name O") map_stats = extract_map_stats_for_single_atoms( pdb_atoms=pdb_atoms, xray_structure=xray_structure, fmodel=fmodel, selection=water_sel) waters = [] for i_seq, atom in enumerate(pdb_atoms) : if (water_sel[i_seq]) : rt_mx_i_inv = asu_mappings.get_rt_mx(i_seq, 0).inverse() self.n_total += 1 asu_dict = asu_table[i_seq] nearest_atom = nearest_contact = None for j_seq, j_sym_groups in asu_dict.items() : atom_j = pdb_atoms[j_seq] site_j = sites_frac[j_seq] # Filter out hydrogens if atom_j.element.upper().strip() in ["H", "D"]: continue for j_sym_group in j_sym_groups: rt_mx = rt_mx_i_inv.multiply(asu_mappings.get_rt_mx(j_seq, j_sym_group[0])) site_ji = rt_mx * site_j site_ji_cart = xray_structure.unit_cell().orthogonalize(site_ji) vec_i = col(atom.xyz) vec_ji = col(site_ji_cart) dxyz = abs(vec_i - vec_ji) if (nearest_contact is None) or (dxyz < nearest_contact) : nearest_contact = dxyz nearest_atom = atom_info(pdb_atom=atom_j, symop=rt_mx) w = water( pdb_atom=atom, b_iso=adptbx.u_as_b(u_isos[i_seq]), occupancy=occupancies[i_seq], nearest_contact=nearest_contact, nearest_atom=nearest_atom, score=map_stats.two_fofc_ccs[i_seq], fmodel=map_stats.fmodel_values[i_seq], two_fofc=map_stats.two_fofc_values[i_seq], fofc=map_stats.fofc_values[i_seq], anom=map_stats.anom_values[i_seq], n_hbonds=None) # TODO if (w.is_bad_water()) : w.outlier = True self.n_bad += 1 elif (w.is_heavy_atom()) : w.outlier = True self.n_heavy += 1 if (w.outlier) or (collect_all) : self.results.append(w) self.n_outliers = len(self.results)
def exercise_2(): pdb_str = """\ CRYST1 12.000 8.000 12.000 90.02 89.96 90.05 P 1 1 ATOM 39 N ASN A 6 5.514 2.664 4.856 1.00 11.99 N ATOM 40 CA ASN A 6 6.831 2.310 4.318 1.00 12.30 C ATOM 41 C ASN A 6 7.854 2.761 5.324 1.00 13.40 C ATOM 42 O ASN A 6 8.219 3.943 5.374 1.00 13.92 O ATOM 43 CB ASN A 6 7.065 3.016 2.993 1.00 12.13 C ATOM 44 CG ASN A 6 5.961 2.735 2.003 1.00 12.77 C ATOM 45 OD1 ASN A 6 5.798 1.604 1.551 1.00 14.27 O ATOM 46 ND2 ASN A 6 5.195 3.747 1.679 1.00 10.07 N ATOM 47 N TYR A 7 8.292 1.817 6.147 1.00 14.70 N ATOM 48 CA TYR A 7 9.159 2.144 7.299 1.00 15.18 C ATOM 49 C TYR A 7 10.603 2.331 6.885 1.00 15.91 C ATOM 50 O TYR A 7 11.041 1.811 5.855 1.00 15.76 O ATOM 51 CB TYR A 7 9.061 1.065 8.369 1.00 15.35 C ATOM 52 CG TYR A 7 7.665 0.929 8.902 1.00 14.45 C ATOM 53 CD1 TYR A 7 6.771 0.021 8.327 1.00 15.68 C ATOM 54 CD2 TYR A 7 7.210 1.756 9.920 1.00 14.80 C ATOM 55 CE1 TYR A 7 5.480 -0.094 8.796 1.00 13.46 C ATOM 56 CE2 TYR A 7 5.904 1.649 10.416 1.00 14.33 C ATOM 57 CZ TYR A 7 5.047 0.729 9.831 1.00 15.09 C ATOM 58 OH TYR A 7 3.766 0.589 10.291 1.00 14.39 O ATOM 59 OXT TYR A 7 11.358 2.999 7.612 1.00 17.49 O TER ATOM 1 N ASN B 6 1.414 5.113 6.019 1.00 12.99 N ATOM 2 CA ASN B 6 2.720 4.776 5.445 1.00 13.30 C ATOM 3 C ASN B 6 3.763 5.209 6.438 1.00 14.40 C ATOM 4 O ASN B 6 4.125 6.391 6.507 1.00 14.92 O ATOM 5 CB ASN B 6 2.922 5.513 4.131 1.00 13.13 C ATOM 6 CG ASN B 6 1.798 5.250 3.160 1.00 13.77 C ATOM 7 OD1 ASN B 6 1.629 4.129 2.686 1.00 15.27 O ATOM 8 ND2 ASN B 6 1.022 6.266 2.875 1.00 11.07 N ATOM 9 N TYR B 7 4.222 4.248 7.230 1.00 15.70 N ATOM 10 CA TYR B 7 5.113 4.552 8.370 1.00 16.18 C ATOM 11 C TYR B 7 6.547 4.754 7.929 1.00 16.91 C ATOM 12 O TYR B 7 6.964 4.259 6.878 1.00 16.76 O ATOM 13 CB TYR B 7 5.042 3.449 9.417 1.00 16.35 C ATOM 14 CG TYR B 7 3.659 3.296 9.977 1.00 15.45 C ATOM 15 CD1 TYR B 7 2.756 2.398 9.402 1.00 16.68 C ATOM 16 CD2 TYR B 7 3.224 4.098 11.023 1.00 15.80 C ATOM 17 CE1 TYR B 7 1.476 2.267 9.896 1.00 14.46 C ATOM 18 CE2 TYR B 7 1.929 3.975 11.545 1.00 15.33 C ATOM 19 CZ TYR B 7 1.063 3.065 10.959 1.00 16.09 C ATOM 20 OH TYR B 7 -0.207 2.910 11.443 1.00 15.39 O ATOM 21 OXT TYR B 7 7.316 5.408 8.654 1.00 18.49 O END """ pdb_in = iotbx.pdb.hierarchy.input(pdb_string=pdb_str) xrs = pdb_in.input.xray_structure_simple() fc = abs(xrs.structure_factors(d_min=1.5).f_calc()) fc = fc.set_observation_type_xray_amplitude() sigf = flex.double(fc.size(), 0.1) + (fc.data() * 0.03) fc = fc.customized_copy(sigmas=sigf) # and now add twinning fc_twin = fc.twin_data(twin_law='-l,-k,-h', alpha=0.4) flags = fc_twin.generate_r_free_flags(use_lattice_symmetry=True) fmodel = mmtbx.utils.fmodel_simple(f_obs=fc_twin, r_free_flags=flags, xray_structures=[xrs], scattering_table="n_gaussian") assert (fmodel.twin_law is not None) map_stats = real_space_correlation.extract_map_stats_for_single_atoms( xray_structure=xrs, pdb_atoms=pdb_in.hierarchy.atoms(), fmodel=fmodel) sel_cache = pdb_in.hierarchy.atom_selection_cache() sel = sel_cache.selection("chain B") map_stats_2 = real_space_correlation.extract_map_stats_for_single_atoms( xray_structure=xrs, pdb_atoms=pdb_in.hierarchy.atoms(), fmodel=fmodel, selection=sel) map_stats_3 = real_space_correlation.map_statistics_for_atom_selection( atom_selection=sel, fmodel=fmodel)
def exercise_2 () : pdb_str = """\ CRYST1 12.000 8.000 12.000 90.02 89.96 90.05 P 1 1 ATOM 39 N ASN A 6 5.514 2.664 4.856 1.00 11.99 N ATOM 40 CA ASN A 6 6.831 2.310 4.318 1.00 12.30 C ATOM 41 C ASN A 6 7.854 2.761 5.324 1.00 13.40 C ATOM 42 O ASN A 6 8.219 3.943 5.374 1.00 13.92 O ATOM 43 CB ASN A 6 7.065 3.016 2.993 1.00 12.13 C ATOM 44 CG ASN A 6 5.961 2.735 2.003 1.00 12.77 C ATOM 45 OD1 ASN A 6 5.798 1.604 1.551 1.00 14.27 O ATOM 46 ND2 ASN A 6 5.195 3.747 1.679 1.00 10.07 N ATOM 47 N TYR A 7 8.292 1.817 6.147 1.00 14.70 N ATOM 48 CA TYR A 7 9.159 2.144 7.299 1.00 15.18 C ATOM 49 C TYR A 7 10.603 2.331 6.885 1.00 15.91 C ATOM 50 O TYR A 7 11.041 1.811 5.855 1.00 15.76 O ATOM 51 CB TYR A 7 9.061 1.065 8.369 1.00 15.35 C ATOM 52 CG TYR A 7 7.665 0.929 8.902 1.00 14.45 C ATOM 53 CD1 TYR A 7 6.771 0.021 8.327 1.00 15.68 C ATOM 54 CD2 TYR A 7 7.210 1.756 9.920 1.00 14.80 C ATOM 55 CE1 TYR A 7 5.480 -0.094 8.796 1.00 13.46 C ATOM 56 CE2 TYR A 7 5.904 1.649 10.416 1.00 14.33 C ATOM 57 CZ TYR A 7 5.047 0.729 9.831 1.00 15.09 C ATOM 58 OH TYR A 7 3.766 0.589 10.291 1.00 14.39 O ATOM 59 OXT TYR A 7 11.358 2.999 7.612 1.00 17.49 O TER ATOM 1 N ASN B 6 1.414 5.113 6.019 1.00 12.99 N ATOM 2 CA ASN B 6 2.720 4.776 5.445 1.00 13.30 C ATOM 3 C ASN B 6 3.763 5.209 6.438 1.00 14.40 C ATOM 4 O ASN B 6 4.125 6.391 6.507 1.00 14.92 O ATOM 5 CB ASN B 6 2.922 5.513 4.131 1.00 13.13 C ATOM 6 CG ASN B 6 1.798 5.250 3.160 1.00 13.77 C ATOM 7 OD1 ASN B 6 1.629 4.129 2.686 1.00 15.27 O ATOM 8 ND2 ASN B 6 1.022 6.266 2.875 1.00 11.07 N ATOM 9 N TYR B 7 4.222 4.248 7.230 1.00 15.70 N ATOM 10 CA TYR B 7 5.113 4.552 8.370 1.00 16.18 C ATOM 11 C TYR B 7 6.547 4.754 7.929 1.00 16.91 C ATOM 12 O TYR B 7 6.964 4.259 6.878 1.00 16.76 O ATOM 13 CB TYR B 7 5.042 3.449 9.417 1.00 16.35 C ATOM 14 CG TYR B 7 3.659 3.296 9.977 1.00 15.45 C ATOM 15 CD1 TYR B 7 2.756 2.398 9.402 1.00 16.68 C ATOM 16 CD2 TYR B 7 3.224 4.098 11.023 1.00 15.80 C ATOM 17 CE1 TYR B 7 1.476 2.267 9.896 1.00 14.46 C ATOM 18 CE2 TYR B 7 1.929 3.975 11.545 1.00 15.33 C ATOM 19 CZ TYR B 7 1.063 3.065 10.959 1.00 16.09 C ATOM 20 OH TYR B 7 -0.207 2.910 11.443 1.00 15.39 O ATOM 21 OXT TYR B 7 7.316 5.408 8.654 1.00 18.49 O END """ pdb_in = iotbx.pdb.hierarchy.input(pdb_string=pdb_str) xrs = pdb_in.input.xray_structure_simple() fc = abs(xrs.structure_factors(d_min=1.5).f_calc()) fc = fc.set_observation_type_xray_amplitude() sigf = flex.double(fc.size(), 0.1) + (fc.data() * 0.03) fc = fc.customized_copy(sigmas=sigf) # and now add twinning fc_twin = fc.twin_data(twin_law='-l,-k,-h', alpha=0.4) flags = fc_twin.generate_r_free_flags(use_lattice_symmetry=True) fmodel = mmtbx.utils.fmodel_simple( f_obs=fc_twin, r_free_flags=flags, xray_structures=[xrs], scattering_table="n_gaussian") assert (fmodel.twin_law is not None) map_stats = real_space_correlation.extract_map_stats_for_single_atoms( xray_structure=xrs, pdb_atoms=pdb_in.hierarchy.atoms(), fmodel=fmodel) sel_cache = pdb_in.hierarchy.atom_selection_cache() sel = sel_cache.selection("chain B") map_stats_2 = real_space_correlation.extract_map_stats_for_single_atoms( xray_structure=xrs, pdb_atoms=pdb_in.hierarchy.atoms(), fmodel=fmodel, selection=sel) map_stats_3 = real_space_correlation.map_statistics_for_atom_selection( atom_selection=sel, fmodel=fmodel)