{+ file: bgroup_twin.inp +} {+ directory: xtal_twin +} {+ description: Grouped, 2-groups per-residue, unrestrained B-factor refinement for data with hemihedral twinning +} {+ authors: Axel T. Brunger, and Paul D. Adams +} {+ copyright: Yale University +} {+ reference: T.O. Yeates, Detecting and Overcoming Crystal Twinning, Methods in Enzymology 276, 344-358 (1997) +} {+ reference: A.T. Brunger, The Free R Value: a Novel Statistical Quantity for Assessing the Accuracy of Crystal Structures, Nature 355, 472-474 (1992) +} {- Guidelines for using this file: - all strings must be quoted by double-quotes - logical variables (true/false) are not quoted - do not remove any evaluate statements from the file -} {========================== IMPORTANT NOTE ===========================} {* This input file requires that the test set has been generated with the CNS task file make_cv_twin.inp. This ensures that twin related reflections are part of the same set. *} {- begin block parameter definition -} define( {======================= molecular structure =========================} {* molecular topology file *} {===>} structure_infile="porin.mtf"; {* parameter files *} {===>} parameter_infile_1="CNS_TOPPAR:protein_rep.param"; {===>} parameter_infile_2=""; {===>} parameter_infile_3=""; {===>} parameter_infile_4=""; {===>} parameter_infile_5=""; {* coordinate file *} {===>} coordinate_infile="porin.pdb"; {====================== crystallographic data ========================} {* space group *} {* use International Table conventions with subscripts substituted by parenthesis *} {===>} sg="R3"; {* unit cell parameters in Angstroms and degrees *} {+ table: rows=1 "cell" cols=6 "a" "b" "c" "alpha" "beta" "gamma" +} {===>} a=104.400; {===>} b=104.400; {===>} c=124.250; {===>} alpha=90; {===>} beta=90; {===>} gamma=120; {* anomalous f' f'' library file *} {* If a file is not specified, no anomalous contribution will be included *} {+ choice: "CNS_XRAYLIB:anom_cu.lib" "CNS_XRAYLIB:anom_mo.lib" "" user_file +} {===>} anom_library=""; {* reflection files *} {* specify non-anomalous reflection files before anomalous reflection files. *} {* files must contain unique array names otherwise errors will occur *} {===>} reflection_infile_1="porin.cv"; {===>} reflection_infile_2=""; {===>} reflection_infile_3=""; {* reciprocal space array containing observed amplitudes: required *} {===>} obs_f="fobs"; {* reciprocal space array containing sigma values for amplitudes: required *} {===>} obs_sigf="sigma"; {* reciprocal space array containing test set for cross-validation: required *} {* cross-validation should always be used, with the possible exception of a final round of refinement including all data *} {* cross-validation is always required for the maximum likelihood targets *} {===>} test_set="test"; {* number for selection of test reflections: required for cross-validation *} {* ie. reflections with the test set array equal to this number will be used for cross-validation, all other reflections form the working set *} {===>} test_flag=1; {* reciprocal space array containing weighting scheme for observed amplitudes: optional *} {* only used for the "residual" and "vector" targets - this will default to a constant value of 1 if array is not present *} {===>} obs_w=""; {* reciprocal space array containing observed intensities: optional *} {* required for the "mli" target *} {===>} obs_i=""; {* reciprocal space array containing sigma values for intensities: optional *} {* required for the "mli" target *} {===>} obs_sigi=""; {* reciprocal space arrays with experimental phase probability distribution: optional *} {* Hendrickson-Lattman coefficients A,B,C,D *} {* required for the "mlhl" target *} {+ table: rows=1 "HL coefficients" cols=4 "A" "B" "C" "D" +} {===>} obs_pa=""; {===>} obs_pb=""; {===>} obs_pc=""; {===>} obs_pd=""; {* complex reciprocal space array containing experimental phases: optional *} {* required for the "mixed" and "vector" targets *} {===>} obs_phase=""; {* reciprocal space array containing experimental figures of merit: optional *} {* required for the "mixed" target *} {===>} obs_fom=""; {* resolution limits to be used in refinement *} {* the full resolution range of observed data should be used in refinement. A bulk solvent correction should be applied to allow the use of low resolution terms. If no bulk solvent correction is applied, data must be truncated at a lower resolution limit of between 8 and 6 Angstrom. *} {+ table: rows=1 "resolution" cols=2 "lowest" "highest" +} {===>} low_res=500.0; {===>} high_res=2.25; {* apply rejection criteria to amplitudes or intensities *} {+ choice: "amplitude" "intensity" +} {===>} obs_type="amplitude"; {* Observed data cutoff criteria: applied to amplitudes or intensities *} {* reflections with magnitude(Obs)/sigma < cutoff are rejected. *} {===>} sigma_cut=0.0; {* rms outlier cutoff: applied to amplitudes or intensities *} {* reflections with magnitude(Obs) > cutoff*rms(Obs) will be rejected *} {===>} obs_rms=10000; {=================== non-crystallographic symmetry ===================} {* NCS-restraints/constraints file *} {* see auxiliary/ncs.def *} {===>} ncs_infile=""; {============ initial B-factor and bulk solvent corrections ==========} {* overall B-factor correction *} {+ choice: "no" "isotropic" "anisotropic" +} {===>} bscale="anisotropic"; {* bulk solvent correction *} {* a mask is required around the molecule(s). The region outside this mask is the solvent region *} {+ choice: true false +} {===>} bulk_sol=true; {* bulk solvent mask file *} {* mask will be read from O type mask file if a name is given otherwise calculated from coordinates of selected atoms *} {===>} bulk_mask_infile=""; {* automatic bulk solvent parameter search *} {+ choice: true false +} {===>} sol_auto=true; {* optional file with a listing of the results of the automatic bulk solvent grid search *} {===>} sol_output=""; {* fixed solvent mask parameters if the automatic option is not used *} {+ table: rows=1 "bulk solvent" cols=2 "probe radius (A)" "shrink radius (A)" +} {===>} sol_rad=1.0; {===>} sol_shrink=1.0; {* fixed solvent parameters if the automatic option is not used *} {+ table: rows=1 "bulk solvent" cols=2 "e-density level (e/A^3)" "B-factor (A^2)" +} {===>} sol_k=-1; {===>} sol_b=-1; {======================= twinning parameters =========================} {* twinning operation *} {===>} twin_oper="h,-h-k,-l"; {* twinning fraction *} {===>} twin_frac=0.304; {========================== atom selection ===========================} {* select atoms to be included in refinement *} {===>} atom_select=(known and not hydrogen); {* select atoms whose B-factor will remain fixed *} {===>} atom_fixed=(none); {* B-factor groups *} {* select atoms in group 1 - protein mainchain example *} {===>} bg_group_1=((resname ALA or resname ARG or resname ASN or resname ASP or resname CYS or resname GLN or resname GLU or resname GLY or resname HIS or resname ILE or resname LEU or resname LYS or resname MET or resname MSE or resname PHE or resname PRO or resname SER or resname THR or resname TRP or resname TYR or resname VAL or resname HISH or resname HISU or resname M3L or resname CPR) and (name c or name n or name ca or name o or name ot+)); {* select atoms in group 2 - protein sidechain example *} {===>} bg_group_2=((resname ALA or resname ARG or resname ASN or resname ASP or resname CYS or resname GLN or resname GLU or resname GLY or resname HIS or resname ILE or resname LEU or resname LYS or resname MET or resname MSE or resname PHE or resname PRO or resname SER or resname THR or resname TRP or resname TYR or resname VAL or resname HISH or resname HISU or resname M3L or resname CPR) and not(name c or name n or name ca or name o or name ot+)); {* select atoms in group 3 - nucleic acid backbone example *} {===>} bg_group_3=((resname THY or resname CYT or resname GUA or resname ADE or resname URI) and (name c+' or name h+' or name op or name o+' or name p or name ht or name ot)); {* select atoms in group 4 - nucleic acid bases example *} {===>} bg_group_4=((resname THY or resname CYT or resname GUA or resname ADE or resname URI) and not(name c+' or name h+' or name op or name o+' or name p or name ht or name ot)); {* select atoms in group 5 - carbohydrates example *} {===>} bg_group_5=(resname GLC or resname GAL or resname MAN or resname NAG or resname FUC or resname SIA or resname XYL); {* select atoms in group 6 *} {===>} bg_group_6=(none); {* select atoms in group 7 *} {===>} bg_group_7=(none); {* select atoms in group 8 *} {===>} bg_group_8=(none); {================= B-factor minimization parameters ==================} {* reset all atomic B factors to this number if positive *} {===>} reset_b=-1; {* number of steps of group B-factor minimization *} {===>} bfactor_nstep=30; {* B-factor limits *} {+ table: rows=1 "B-factor" cols=2 "minimum" "maximum" +} {===>} bmin=1; {===>} bmax=200; {* refinement target *} {+ list: twin_lsq: least squares residual for hemihedral twinning +} {+ choice: "twin_lsq" +} {===>} reftarget="twin_lsq"; {* number of bins for refinement target *} {* this will be determined automatically if a negative value is given otherwise the specified number of bins will be used *} {===>} target_bins=-1; {* memory allocation for FFT calculation *} {* this will be determined automatically if a negative value is given otherwise the specified number of words will be allocated *} {===>} fft_memory=-1; {=========================== output files ============================} {* output coordinate file *} {===>} coordinate_outfile="bgroup_twin.pdb"; {* format output coordinates for use in o *} {* if false then the default CNS output coordinate format will be used *} {+ choice: true false +} {===>} pdb_o_format=true; {===========================================================================} { things below this line do not normally need to be changed } {===========================================================================} ) {- end block parameter definition -} checkversion 1.2 evaluate ($log_level=quiet) structure @&structure_infile end coordinates @&coordinate_infile parameter if ( &BLANK%parameter_infile_1 = false ) then @@¶meter_infile_1 end if if ( &BLANK%parameter_infile_2 = false ) then @@¶meter_infile_2 end if if ( &BLANK%parameter_infile_3 = false ) then @@¶meter_infile_3 end if if ( &BLANK%parameter_infile_4 = false ) then @@¶meter_infile_4 end if if ( &BLANK%parameter_infile_5 = false ) then @@¶meter_infile_5 end if end xray @CNS_XTALLIB:spacegroup.lib (sg=&sg; sgparam=$sgparam;) a=&a b=&b c=&c alpha=&alpha beta=&beta gamma=&gamma @CNS_XRAYLIB:scatter.lib if ( &BLANK%reflection_infile_1 = false ) then reflection @@&reflection_infile_1 end end if if ( &BLANK%reflection_infile_2 = false ) then reflection @@&reflection_infile_2 end end if if ( &BLANK%reflection_infile_3 = false ) then reflection @@&reflection_infile_3 end end if end if ( &BLANK%anom_library = false ) then @@&anom_library else set echo=off end xray anomalous=? end if ( $result = true ) then display Warning: no anomalous library has been specified display no anomalous contribution will used in refinement end if set echo=on end end if set echo=off end if ( &twin_frac > 0.5 ) then display Error: twinning fraction must be less than or equal 0.5 abort end if set echo=on end {- copy define parameters of optional arrays into symbols so we can redefine them -} evaluate ($obs_i=&obs_i) evaluate ($obs_sigi=&obs_sigi) evaluate ($obs_w=&obs_w) xray @@CNS_XTALMODULE:checkrefinput ( reftarget=&reftarget; obs_f=&obs_f; obs_sigf=&obs_sigf; test_set=&test_set; obs_pa=&obs_pa; obs_pb=&obs_pb; obs_pc=&obs_pc; obs_pd=&obs_pd; obs_phase=&obs_phase; obs_fom=&obs_fom; obs_w=$obs_w; obs_i=$obs_i; obs_sigi=$obs_sigi; ) query name=fcalc domain=reciprocal end if ( $object_exist = false ) then declare name=fcalc domain=reciprocal type=complex end end if declare name=fbulk domain=reciprocal type=complex end do (fbulk=0) ( all ) binresolution &low_res &high_res mapresolution &high_res if ( &obs_type = "intensity" ) then if ( &BLANK%obs_i = true ) then display Error: observed intensity array is undefined display aborting script abort end if evaluate ($reject_obs=&obs_i) evaluate ($reject_sig=&obs_sigi) show min (amplitude(&STRIP%obs_i)) (all) evaluate ($obs_lower_limit=$result-0.1) else evaluate ($reject_obs=&obs_f) evaluate ($reject_sig=&obs_sigf) evaluate ($obs_lower_limit=0) end if declare name=ref_active domain=reciprocal type=integer end declare name=tst_active domain=reciprocal type=integer end do (ref_active=0) ( all ) do (ref_active=1) ( ( amplitude($STRIP%reject_obs) > $obs_lower_limit ) and ( &low_res >= d >= &high_res ) ) statistics overall completeness selection=( ref_active=1 ) end evaluate ($total_compl=$expression1) show sum(1) ( ref_active=1 ) evaluate ($total_read=$select) evaluate ($total_theor=int(1./$total_compl * $total_read)) show rms (amplitude($STRIP%reject_obs)) ( ref_active=1 ) evaluate ($obs_high=$result*&obs_rms) show min (amplitude($STRIP%reject_obs)) ( ref_active=1 ) evaluate ($obs_low=$result) do (ref_active=0) ( all ) do (ref_active=1) ( ( amplitude($STRIP%reject_obs) >= &sigma_cut*$STRIP%reject_sig ) and ( $STRIP%reject_sig # 0 ) and ( $obs_low <= amplitude($STRIP%reject_obs) <= $obs_high ) and ( $obs_low <= amplitude(remap[&STRIP%twin_oper]($STRIP%reject_obs)) <= $obs_high ) and ( &low_res >= d >= &high_res ) ) do (tst_active=0) (all) if ( &BLANK%test_set = false ) then do (tst_active=1) (ref_active=1 and &STRIP%test_set=&test_flag) end if show sum(1) ( ref_active=1 and tst_active=0 ) evaluate ($total_work=$select) show sum(1) ( ref_active=1 and tst_active=1 ) evaluate ($total_test=$select) evaluate ($total_used=$total_work+$total_test) evaluate ($unobserved=$total_theor-$total_read) evaluate ($rejected=$total_read-$total_used) evaluate ($per_unobs=100*($unobserved/$total_theor)) evaluate ($per_reject=100*($rejected/$total_theor)) evaluate ($per_used=100*($total_used/$total_theor)) evaluate ($per_work=100*($total_work/$total_theor)) evaluate ($per_test=100*($total_test/$total_theor)) associate fcalc ( &atom_select ) tselection=( ref_active=1 ) cvselection=( tst_active=1 ) method=FFT {- MODIFIED 8/01/06 -} end show min ( b ) ( &atom_select ) evaluate ($b_min=$result) @@CNS_XTALMODULE:fft_parameter_check ( d_min=&high_res; b_min=$b_min; grid=auto; fft_memory=&fft_memory; fft_grid=$fft_grid; fft_b_add=$fft_b_add; fft_elim=$fft_elim; ) xray {- END MODIFICATION -} tolerance=0.0 lookup=false end igroup interaction (&atom_select) (&atom_select) end flags exclude * include xref end if ( &BLANK%ncs_infile = false ) then inline @&ncs_infile end if if ( &reset_b > 0 ) then do (b=&reset_b) (&atom_select) end if xray predict mode=reciprocal to=fcalc selection=(ref_active=1) atomselection=( &atom_select ) end end {- BEGIN MODIFICATION 8/01/06 -} @CNS_XTALMODULE:scale_and_solvent_grid_search ( bscale=&bscale; sel=( ref_active=1 ); sel_test=( tst_active=1 ); atom_select=( &atom_select ); bulk_sol=&bulk_sol; bulk_mask=&bulk_mask_infile; bulk_atoms=( &atom_select ); sol_auto=&sol_auto; sol_k=&sol_k; sol_b=&sol_b; sol_rad=&sol_rad; sol_shrink=&sol_shrink; fcalc=fcalc; obs_f=&STRIP%obs_f; obs_sigf=&STRIP%obs_sigf; obs_i=$STRIP%obs_i; obs_sigi=$STRIP%obs_sigi; fpart=fbulk; Baniso_11=$Baniso_11; Baniso_22=$Baniso_22; Baniso_33=$Baniso_33; Baniso_12=$Baniso_12; Baniso_13=$Baniso_13; Baniso_23=$Baniso_23; Biso=$Biso_model; sol_k_best=$sol_k_ref; sol_b_best=$sol_b_ref; solrad_best=$solrad_best; shrink_best=$shrink_best; b=b; low_b_flag=$low_b_flag; sol_output=&sol_output; ) {- check the gridding again since the minimum B-factor may have changed -} show min ( b ) ( &atom_select ) evaluate ($b_min=$result) @@CNS_XTALMODULE:fft_parameter_check ( d_min=&high_res; b_min=$b_min; grid=auto; fft_memory=&fft_memory; fft_grid=$fft_grid; fft_b_add=$fft_b_add; fft_elim=$fft_elim; ) {- END MODIFICATION -} xray @@CNS_XTALMODULE:refinementtarget_twin (target=&reftarget; sig_sigacv=0.07; mbins=&target_bins; fobs=&STRIP%obs_f; sigma=&STRIP%obs_sigf; weight=$STRIP%obs_w; iobs=$STRIP%obs_i; sigi=$STRIP%obs_sigi; test=tst_active; fcalc=fcalc; fpart=fbulk; twin_oper=&STRIP%twin_oper; twin_frac=&twin_frac; pa=&STRIP%obs_pa; pb=&STRIP%obs_pb; pc=&STRIP%obs_pc; pd=&STRIP%obs_pd; phase=&STRIP%obs_phase; fom=&STRIP%obs_fom; sel=(ref_active=1); sel_test=(tst_active=1); statistics=true;) end @@CNS_XTALMODULE:one_term_wa (wa=$wa;) xray @@CNS_XTALMODULE:calculate_r_twin (fobs=&STRIP%obs_f; fcalc=fcalc; fpart=fbulk; twin_oper=&STRIP%twin_oper; twin_frac=&twin_frac; sel=(ref_active=1); sel_test=(tst_active=1); print=true; output=OUTPUT; r=$start_r; test_r=$start_test_r;) end evaluate ($ngroup=1) evaluate ($group=1) evaluate ($done=false) while ( $done = false ) loop group if ( &exist_bg_group_$group = true ) then show sum(1) ( &bg_group_$group ) evaluate ($size_$group=$result) evaluate ($ngroup=$ngroup+1) else evaluate ($done=true) evaluate ($ngroup=$ngroup-1) end if evaluate ($group=$group+1) end loop group xray optimize group bmin=&bmin bmax=&bmax evaluate ($counter=1) while ( $counter <= $ngroup ) loop group if ( $size_$counter > 0 ) then for $id in id (tag) loop resid b=(((&bg_group_$counter) and byres id $id) and not(&atom_fixed)) end loop resid end if evaluate ($counter=$counter+1) end loop group nstep=&bfactor_nstep drop=1.0 ? end end xray predict mode=reciprocal to=fcalc selection=(ref_active=1) atomselection=( &atom_select ) end @@CNS_XTALMODULE:calculate_r_twin (fobs=&STRIP%obs_f; fcalc=fcalc; fpart=fbulk; twin_oper=&STRIP%twin_oper; twin_frac=&twin_frac; sel=(ref_active=1); sel_test=(tst_active=1); print=true; output=OUTPUT; r=$full_r; test_r=$full_test_r;) end set display=&coordinate_outfile end display REMARK coordinates from twinned data group B-factor refinement display REMARK twinning operator= &STRIP%twin_oper twinning fraction= &twin_frac display REMARK refinement resolution: &low_res - &high_res A if ( $total_test > 0 ) then display REMARK starting twinned r= $start_r[f6.4] twinned free_r= $start_test_r[f6.4] display REMARK final twinned r= $full_r[f6.4] twinned free_r= $full_test_r[f6.4] else display REMARK starting twinned r= $start_r[f6.4] display REMARK final twinned r= $full_r[f6.4] end if xray wa=? end evaluate ($wa_print=$result) display REMARK wa= $wa_print display REMARK target= &STRIP%reftarget steps= &bfactor_nstep display REMARK sg= &STRIP%sg a= &a b= &b c= &c alpha= &alpha beta= &beta gamma= &gamma if ( &BLANK%parameter_infile_1 = false ) then display REMARK parameter file 1 : &STRIP%parameter_infile_1 end if if ( &BLANK%parameter_infile_2 = false ) then display REMARK parameter file 2 : &STRIP%parameter_infile_2 end if if ( &BLANK%parameter_infile_3 = false ) then display REMARK parameter file 3 : &STRIP%parameter_infile_3 end if if ( &BLANK%parameter_infile_4 = false ) then display REMARK parameter file 4 : &STRIP%parameter_infile_4 end if if ( &BLANK%parameter_infile_5 = false ) then display REMARK parameter file 5 : &STRIP%parameter_infile_5 end if display REMARK molecular structure file: &STRIP%structure_infile display REMARK input coordinates: &STRIP%coordinate_infile if ( &BLANK%anom_library = false ) then display REMARK anomalous f' f'' library: &STRIP%anom_library end if if ( &BLANK%reflection_infile_1 = false ) then display REMARK reflection file= &STRIP%reflection_infile_1 end if if ( &BLANK%reflection_infile_2 = false ) then display REMARK reflection file= &STRIP%reflection_infile_2 end if if ( &BLANK%reflection_infile_3 = false ) then display REMARK reflection file= &STRIP%reflection_infile_3 end if if ( &BLANK%ncs_infile = false ) then display REMARK ncs= &STRIP%ncs_type ncs file= &STRIP%ncs_infile else display REMARK ncs= none end if if ( &bscale # "no" ) then if ( $low_b_flag = true ) then display REMARK warning: B-correction gave atomic B-values less than zero display REMARK they have been reset to zero end if end if ! ! Begin modification (8/01/06) if ( &bscale = "anisotropic" ) then display REMARK Anisotropic B-factor tensor Ucart of atomic model without isotropic component : display REMARK B11=$Baniso_11[f8.3] B22=$Baniso_22[f8.3] B33=$Baniso_33[f8.3] display REMARK B12=$Baniso_12[f8.3] B13=$Baniso_13[f8.3] B23=$Baniso_23[f8.3] display REMARK Isotropic component added to coordinate array B: $Biso_model[f8.3] elseif ( &bscale = "isotropic" ) then display REMARK B-factor applied to coordinate array B: $Biso_model[f8.3] else display REMARK initial B-factor correction: none end if if ( &bulk_sol = true ) then display REMARK bulk solvent: probe radius=$solrad_best, shrink value=$solrad_best display REMARK bulk solvent: density level= $sol_k_ref e/A^3, B-factor= $sol_b_ref A^2 else display REMARK bulk solvent: false end if {- END MODIFICATION -} if ( &obs_type = "intensity" ) then display REMARK reflections with Iobs/sigma_I < &sigma_cut rejected display REMARK reflections with Iobs > &obs_rms * rms(Iobs) rejected display REMARK reflections with Iobs[&STRIP%twin_oper] = 0 rejected else display REMARK reflections with |Fobs|/sigma_F < &sigma_cut rejected display REMARK reflections with |Fobs| > &obs_rms * rms(Fobs) rejected display REMARK reflections with |Fobs|[&STRIP%twin_oper] = 0 rejected end if xray anomalous=? end if ( $result = true ) then display REMARK anomalous diffraction data was input end if {- MODIFIED 8/01/06 -} display REMARK fft gridding factor = $fft_grid, B factor offset = $fft_b_add A^2, Elimit = $fft_elim {- END MODIFICATION -} display REMARK theoretical total number of refl. in resol. range: $total_theor[I6] ( 100.0 % ) display REMARK number of unobserved reflections (no entry or |F|=0): $unobserved[I6] ( $per_unobs[f5.1] % ) display REMARK number of reflections rejected: $rejected[I6] ( $per_reject[f5.1] % ) display REMARK total number of reflections used: $total_used[I6] ( $per_used[f5.1] % ) display REMARK number of reflections in working set: $total_work[I6] ( $per_work[f5.1] % ) display REMARK number of reflections in test set: $total_test[I6] ( $per_test[f5.1] % ) remark @CNS_XTALMODULE:write_pdb (pdb_o_format=&pdb_o_format; coordinate_outfile=&coordinate_outfile; sgparam=$sgparam;) stop