X-PLOR

2       Methodology

This chapter provides a brief account of the basic ideas, practical consequences and implementation of the new features in this release of the X-PLOR software for X-ray crystallographic refinement and NMR structure determination.

X-ray crystallographic refinement

By eliminating bond lengths and angles as degrees of freedom from the molecule it is also possible to use much larger time steps for integrating the equations of motion with torsion angle dynamics than is possible with conventional molecular dynamics. The most useful consequence of the ability to run stable refinements with large time steps is that difficult refinement problems, which require simulated annealing at high temperatures in order to escape from incorrect conformations, can be tackled. For conventional molecular dynamics the computational cost of high temperature refinements is prohibitively high because very small time steps are required to control vibrations along the bond lengths and angles.

The application of torsion angle dynamics to X-ray crystallographic refinement has been shown to give somewhat improved results over conventional simulated annealing methods at a given temperature (Rice and Brünger 1994). More significantly, torsion angle refinements starting at very high temperatures give converged refinements for poor starting models -- a result that can not be achieved using conventional simulated annealing protocols confined to lower temperatures.

The code for the torsion angle dynamics algorithm in X-PLOR (Rice and Brünger 1994), released in interim form in X-PLOR 3.851, has now been optimized to give better performance. For a 120-amino-acid protein, the integration of the equations of motion is approximately 40% faster than in the original implementation.

Implementation

The parameters that control the torsion angle dynamics are contained in a dynamic torsion command block, which is terminated with an end statement. The available parameters can be classified into two types (i) parameters that control the initial setup of the torsion angle topology and (ii) parameters that control the dynamics run and which are analogous to parameters available for conventional molecular dynamics.

Specification of topology

The parameters that control the specification of the torsion topology are contained within a topology sub-block which is terminated with an end statement.

The available parameters are:

FIX TORSion <selection> which constrains selected torsions.

FIX GROUp <selection> which constrains all torsions in <selection>.

FREE TORSion <selection> which removes a torsion constraint.

FREE BOND <selection> which ignores a connection for generating the topology.

KDIHmax = <real> which sets a threshold energy for establishing free torsions.

MAXTree = <integer> which sets the maximum number of trees in the structure.

MAXChain = <integer> which sets the maximum number of chains per tree.

MAXLength= <integer> which sets the maximum length of a chain in a tree.

MAXBond = <integer> which sets the maximum number of bonds allowed per atom.

MAXDihe = <integer> which sets the maximum number of torsions allowed per atom.

MAXJoint = <integer> which sets the maximum number of branches per atomic group.

RESET is used to eliminate any previous topology specifications.

The following parameters are supported for specifying the simulated annealing protocol:

ASCIi=<logical> writes a formatted (ASCII) coordinate trajectory file if ASCIi is TRUE; if ASCIi is FALSE an unformatted (binary) file is written.In the former case, a scale factor and an offset are applied to the co-ordinates and the result is converted to an integer number.

CSEL=<selection> selects which atoms are written to the coordinate trajectory file.

ILBFrq=<integer> updates the Langevin boundary between normal and Langevin particles in conjunction with RBUF and ORIGin.

NSTEp=<integer> sets the number of dynamics steps.

NPRInt=<integer> determines the frequency with which the energy information is printed.

NSAVC=<integer> determines the frequency with which the coordinates are written to the trajectory file.

NSAVV=<integer> determines the frequency with which the coordinates are written to the coordinate trajectory file.

OFFSet=<real> provides the offset for the ASCIi=TRUE option.

ORIGin=<vector> is the origin of the sphere of the Langevin boundary.

RBUF=<real> specifies the radius of the spherical boundary in for the normal and Langevin particles.

REASsign=<logical> reassigns the velocities if TRUE.

SCALe=<real> provides a scale factor for the ASCIi=TRUE option.

TBATh=<real> specifies the temperature of the heat bath or the target temperature for the temperature coupling.

TIMEstep=<real> sets the time step for the dynamics.

TRAJectory=<filename> provides the filename of the coordinate trajectory file.

VSEL=<selection> selects atoms that are written to the velocity trajectory file.

VELOcity=<filename> names the velocity trajectory file.

VASCii=<logical> writes a formatted (ASCII) coordinate trajectory file if ASCIi is TRUE; if ASCIi is FALSE an unformatted (binary) file is written.In the former case, a scale factor and an offset are applied to the velocities and the result is converted to an integer number.

VOFFset=<real> provides the offset for the ASCIi=TRUE option.

VSCALe=<real> provides a scale factor for the ASCIi=TRUE option.

Background

E = w(h)(|F_obs(h)|-k|F_calc(h)|)²

This target would give correct results if (i) the deviation between |Fob_s(h)| and k|Fcalc(h) was Gaussian, (ii) the mean deviation was zero and (iii) the standard deviation of the Gaussian was independent of the parameters of the atomic model. This latter point is clearly untrue since the errors have a changing phase component which depends on the atomic model.

To overcome these fundamental difficulties, new targets for crystallographic refinement have been derived from first principles using maximum likelihood statistics (Pannu and Read 1996). The resulting target takes the form

E = w_ml(h) (|F_obs(h)|-<|F_obs(h)|>)²

for refinement against structure factor amplitudes and an analogous expression is obtained for refinement against intensities. Although this target is superficially similar to the least-squares residual, the weight, w_ml, and expectation value of the observed structure factor, <|Fobs(h)|>, are now calculated via estimates of _A (

Practical tests involving minimizations of misfit molecular models with the maximum likelihood targets and the conventional residual target show that better convergence (lower Rfree) and reduced bias (smaller difference between R and R_free) is obtained with the maximum likelihood targets (Pannu and Read 1996). Tests in which a maximum likelihood target was used in conjunction with torsion angle dynamics showed that a greater radius of convergence could be achieved than with other refinement methods (Adams et al. 1997).

More recently, a new maximum likelihood target, the structure factor amplitude with Hendrickson-Lattman phase probability coefficients, has been developed and this target is also available in the X-PLOR program. This target makes optimal use of all the experimental information that is available since the phase probability coefficients correctly model the uncertainty in the experimentally determined values of the phase angle (that is., as determined from MIR or MAD data). Preliminary tests of the maximum likelihood target with phase probability information (Pannu and Read, unpublished results) give very promising results.

Implementation

• MLF1, the amplitude based target

• MLF2, the intensity based target

• MLHL, a target which makes use of Hendrickson-Lattman phase probability coefficients

There are two other parameters within the xrefin target block, siga and mbins, which can alter the behavior of maximum likelihood refinement.

The siga parameter provides the ability to update the _A estimate that is used in the calculation of the maximum likelihood target. The siga parameter may be set to fix, next, or refi=<integer>.

The siga fix option (the default) fixes the estimates for _A at the values used at the beginning of the refinement. To update the estimates for _A after some cycles of refinement the siga next option may be used. The siga refi = <integer> option is used to update the estimates for the _A values after the given number of structure factor calculations. This latter option should be used with caution because frequent updating sometimes causes an increase in E_xref which can cause the line search to be abandoned in minimization.

The mbins parameter sets the number of resolution bins used for estimation of the _A values. The default value of mbins for other calculations with X-PLOR that use this parameter (for example, resolution-dependent tabulation of R-values -- see page 164 of the X-PLOR 3.1 manual) is 8. If maximum likelihood refinement is performed the default value for this parameter will automatically change to the number of reflections divided by 1000 or the number of cross-validation reflections divided by 50, whichever is the smaller. This new default will usually be a good estimate for the optimal value of mbins for the refinement. Any value of mbins that is explicitly set in the X-PLOR script will override these default modes of operation.

As in refinements against the other targets, a weight parameter, wa, is needed to scale the gradients in the X-ray energy to the gradients in the chemical energy. The value of the wa can be estimated in the same way as for the other targets, by using a script (/tutorial/xtalrefine/check.inp) which carries out a short free dynamics run and then calculates the ratios of the chemical and X-ray energy gradients. Since the maximum likelihood refinements make use of (internally) normalized data, the correct value of wa is usually orders of magnitude smaller than the value that would be used for a refinement with the least-squares residual. Note that none of the maximum likelihood targets uses the phase weight parameter, wp, (the MLHL refinement target deals directly with phase probabilities, not any explicit value for a phase angle) so wp should either be set to zero or omitted from the refinement script.

It should be noted that a set of cross-validation data is an absolute requirement for maximum likelihood refinement. If the cross-validation reflections are not flagged in the input data file and are not explicitly set up in the refinement script, 10% of the data will be automatically used for cross-validation purposes.

Example scripts for refinement using the maximum likelihood targets are /tutorial/xtaltorsion/torsion_slow_ml.inp for co-ordinate refinements and /tutorial/xtalrefine/brefinement_ml.inp for temperature factor refinements.

Andersen thermal coupling

X-PLOR now also offers the option of temperature control by the Andersen thermal coupling method (Andersen 1980). In the Andersen method a set of atoms is randomly selected from the molecule and their velocities are replaced by randomly generated velocities selected from a Boltzmann velocity distribution at the desired temperature. The physical process that is modeled by the Andersen thermal coupling algorithm is a collision of a particle in the system with a particle in a heat bath at the desired temperature.

In contrast to the Berendsen thermal coupling method, which simply modifies the trajectories of all atoms in the system by rescaling along their current paths, the Andersen thermal coupling involves a stochastic process which alters the velocities and directions of the selected atoms at each time step. The general effect of Andersen thermal coupling is to break up correlated motions involving groups of atoms. This behavior increases local sampling of the conformational space but inhibits more global changes in the protein model.

Practical tests of the Andersen thermal coupling method reflect these theoretical expectations. For atomic models that are fairly close to convergence the Andersen thermal coupling method usually gives final structures with slightly lower values of R_free than structures obtained from equivalent calculations with the Berendsen thermal coupling method. Models that still require large concerted movements to bring them into agreement with the X-ray data tend to be less convergent (unpublished results). These findings suggest that the Andersen thermal coupling method might be usefully employed in the final stages of a simulated annealing refinement. For example, in a slow cooling refinement protocol, where the temperature is reduced from 3000 to 300K, one could switch to the Andersen thermal coupling method once the system reached 1000K.

Implementation

The Andersen thermal coupling option is turned on by setting andersen = true (default is false). A control parameter, nequ=<integer>, specifies the number of time steps needed to reset the velocities of N_atom atoms, where N_atom is the number of atoms in the molecule. As a rough guide, for a cooling stage involving N time steps, nequ should be set to about N/2 to achieve full temperature control. A typical value for nequ would be 20.

Note that the Berendsen and Andersen thermal coupling options are mutually exclusive so the option to turn on Berendsen thermal coupling, tcoupling=true, should not be used if the Andersen thermal coupling option is used.

MAD phasing

Background

Sequence of jobs to be run

Merges reflection files.

Scales data.

Computes dispersive and anomalous diffraction ratios.

Computes histograms of dispersive and anomalous differences.

MAD heavy atom refinement and computation of phase probability distributions.

Compute fom-weighted electron density map from MAD phases.

Compute fom-weighted dispersive difference Fourier map between wavelengths.

Compute fom-weighted anomalous difference Fourier map at a wavelength.

You should restrict the resolution limits of the merged data to the range that you want to use in heavy atom searches and refinements. Writing all data without actually using it will cause additional CPU expense although the result will be identical.

In the output reflection file produced by mad_merge.inp the structure factor amplitudes are named F_W1, F_W2, F_W3,.... and the corresponding standard deviations are named S_W1, S_W2,S_W3,...

3.   Edit and run the mad_scale.inp file.

Modify the number of atoms according to the expected asymmetric unit contents. Also modify the spacegroup, unit cell, etc. Then run mad_scale.inp.

The anomalous and dispersive diffraction ratios will be listed in the output file, mad_analyse.dat.

The anomalous and dispersive histograms will be listed in the output file, mad_histogram.dat.

Macrocycles, cycles, and refinement steps

The first macrocycle is always carried out without refinement steps and scaling. Refinement steps and scaling start in the second macrocycle. This is done in order to build up an initial combined phase probability distribution using all available information before performing the refinement steps and scaling. At the end of each cycle of a particular LOC, the overall combined probability phase distribution is updated.

LOCs

F->F_ref and Friedel(F)->F_ref.

For the reference wavelength only the Friedel(F_ref)->F_ref LOC is used. Refinement of the LOCs is carried out independently constrained by the overall phase probability distribution. Differences in refinement parameters among the various LOCs can be used to assess the convergence of the refinement.

Phasing and refinement selections

Cross-validation

Compressing the log file

Assessing convergence

Note that in the log file the "cycle" number refers to the number of microcycles. Note that the fp and fdp shifts are identical for all sites at a particular wavelength whereas B values and coordinates are individually refined. Some experimentation with $xstep, $bstep, $fpstep, $fdpstep, $macrocycle, $ncycle, $phistep, $tolerance should be performed to obtain a reasonable compromise between computer time and accuracy.

Fp and fdp shifts are the most sensitive indicators of convergence. One can get these shifts by using a pipeline of grep commands.

For example:

grep TAB mad_long.out | grep shift | grep "F_W1" | grep "1 SITE" | grep fp

If the shifts are below 0.1 the corresponding phases should have converged and one can stop the refinement.

Other indicators for convergence are the difference in f' and f'' values for the F->F_ref and Friedel(F)->F_ref LOCs (cf. file mad_refine.fp), the difference in coordinate and B-factor values for the various LOCs (cf. file mad_refine.pdb), reasonable final B values and coordinates and shifts with respect to the initial parameters (cf. file mad_refine.summary).

Choice of refinement and phasing parameters

The order in which the LOCs are computed should produce only a small influence on the result (the "list" parameter in the macro call to master). The choice of reference wavelength ($ref_wave) can produce slightly different phases. Thus, the phase differences between different choices of reference wavelengths are a lower bound of the phase error. It is sometimes also of interest to determine the influence of omitting a particular wavelength ($wave_1, $wave_2,... flags).

Lack-of-closure expressions

The individual LOCs can be custom-tailored (that is, different selections for the reflections used for the target and scale calculations, refinement parameters) by changing the entries in the call to the master macro.

Absolute scale of structure factors and scattering factors

f' (new) = scale * f'(refined) + offset

f'' (new) = scale * f''(refined)

fobs(new) = scale * fobs(refined)

sigma(new) = scale * sigma(refined)

f_w1(new) = scale * f_w1(refined)

s_w1(new) = scale * s_w1(refined)

...

scale = f''(initial) / f''(refined)

for "scaling" wavelength Friedel(F)->F_ref LOC

offset= f'(initial) - scale*f'(refined)

for "scaling" wavelength Friedel(F)->F_ref LOC

Interpretation of the output files produces by mad_refine.inp

This file contains the refined f' and f'' values. Note that the probabilistic approach cannot produce absolute values for f' and f''. The scale of f' and f'' is dependent on the scale of the diffraction data (the scale factor that was applied to the reference wavelength by mad_scale.inp). In addition, f' is always relative to a baseline which has to be theoretically estimated from the observed fluorescence signal. If the "scaling" wavelength is specified, f' and f'' and offset and re-scaled to match the scaling wavelength. The refinement is performed independently for the F -> F_ref and Friedel(F) -> F_ref LOCs. This provides a way of assessing the convergence of the refinement. According to Brünger, the corresponding values should normally be close (within 10% of the maximum f' and f'' values).

Note:

1. The F_ref-F_ref LOC is not meaningful, thus no refined f', f'' values are given.

2. If no "scaling" wavelength is available, the f' value for the reference wavelength will be zero and the scale of f' and f'' will be determined by the scaling performed in mad_scale.inp.

3. The scaling wavelength does not have to be equal to the reference wavelength.

This file contains the phase probability distributions (stored in arrays PA, PB, PC, PD), the corresponding phase centroids (stored in the phases of the FOBS array), and the corresponding figures-of-merit. This file is ready to be read by subsequent jobs to produce experimental electron density maps, perform density modification, phase combination, and refinement with or without phase information.

3.   mad_refine.pdb

This file contains the refined coordinates of the anomalous sites. There is one coordinate set for each LOC, that is, each LOC is treated independently. The SEGIDs are encoded according to the particular LOC, for example:

The SEGID TO+2 refers to the F->Fref LOC and the SEGID TO-2 refers to the Friedel(F)->F_ref LOC.

Convergence of the coordinate sets and B factors for corresponding sites can be used to assess the quality of the refinement. According to Brünger, the coordinates are very similar for the different LOCs. The B-values show more variation since they may compensate for systematic errors between different wavelengths.

4.   mad_refine.summary

This file lists the LOC R value, LOC value, phasing power, initial and final coordinates and other refined parameters. The overall figure-of-merit distribution is listed at the end.

Statistical indicators for the working set

1. Phasing power values > 1.0: excellent LOC

2. R-Cullis < 0.6: excellent LOC;
   R-Cullis < 0.9: usable LOC

write reflection fobs sigma

Release notes and warnings

1. At the current stage of the program, individual heavy atom directions cannot be fixed during refinement: the heavy atom is either refined or kept fixed. This may cause drifting of the heavy atom in cases where one of the directions is arbitrary (for example, for one heavy atom in P2₁, the y-coordinate is arbitrary).

2. At the current stage of the program the Hendrickson-Lattman coefficients are treated as real arrays. This means that if the data set is expanded or mapped into a different asymmetric unit the corresponding phase probability distributions may be wrong. To expand and reduce Hendrickson-Lattman coefficient data sets use the expandprobability and reduceprobability macros.

3. You must reduce all your data sets in anomalous mode: Friedel mates are kept as separate reflections. X-PLOR will phase both Friedel mates. To obtain phases for the non-anomalous structure factor follow the note below.

4. Please make sure you run enough cycles to get good convergence (see notes below).

5. Note that the individual FOMs listed in the output file (for example, when using the grep command to get the lines marked by "TAB") represent the FOMs for each lack-of-closure expression. The FOMs of the combined phase probability distributions are listed in the mad_refine_summary file.

6. Note that the notion of a "test" set in the macros should not be confused with a randomly selected set. In this context it refers to the reflections that are not used for scaling and refinement but that are phased.

7. The refinement can be very slow because of the generality of the method.

This utility program is able to read a variety of data types and arrangements. If the data is initially in a binary format (for example, as in the CCP4 data formats) you should dump these to an ASCII format (CCP4 provides a facility for this operation) and then select the appropriate format for this utility program.

Implementation

xplor_data_convert < data_convert.inp > data_convert.log

The input script contains the following information:

1. Crystal cell dimensions (a, b, c, , , )

2. Lower and upper resolution limits in angstroms.

3. Codes to specify the input data items. Reflection records are assumed to begin with Miller indices h,k,l. The data items are read as free-format items:

FSFS F_obs(+), SD(+), F_obs(-), SD(-)

FSD F_obs, SD, Anomalous difference

FSFP F_obs, SD, F_calc, Phase_calc

FSPM F_obs, SD, Phase, Figure-of-merit

FSMP F_obs, SD, Figure-of-merit, Phase

FMP F_obs, Figure-of-merit, Phase

FS F_obs, SD

IS Intensity, SD

ISIS Intensity(+), SD(+), Intensity(-), SD(-)

FMP4 F_obs, Figure-of-merit, Phase, A, B, C, D

FSMP4 F_obs, SD, Figure-of-merit, Phase, A, B, C, D

1. An option (yes/no) to use a FORTRAN-style format description for the input reflection data. This option provides the capability of reading files that do not precisely match one of the pre-set format descriptions on line 3. For example, the format descriptor could be used to bypass unwanted items.

2. A FORTRAN-style format descriptor (read but not interpreted if "no" is entered on line 4).

3. Specification (yes/no) as to whether the input data set contains Bijvoet pairs as unique items or whether these data have been merged.

NMR structure determination

In summary, this protocol consists of the following stages:

1. `Cooking' at 50,000K for 15ps with the energy constant for the Van der Waals parameters scaled by 0.1.

2. Cooling to 1000K for 15ps with ramping of the Van der Waals parameters to full scale.

3. Cooling to 300K for 6ps using conventional molecular dynamics.

4. 1000 steps of conjugate gradient minimization.

The weights for the NOE restraints are set to 150 kcal/mol for stages 1-3 and 50 kcal/mol for stage 4. The functional form for the NOE distance restraints is a flat-bottomed parabolic function with a soft asymptote. The weights for the dihedral angle restraints are set to 100 kcal/mol for stages 1-3 and 300 kcal/mol in the final stage.

Implementation

Structure determination using ambiguous restraints

Introduction

Each task performed by ARIA has a corresponding subdirectory containing example scripts in /tutorial/aria/):

calibrate peak volumes to distance restraints, based on reference volumes and distances, or calculated structures. Several proton classes can be used.

assign ambiguous NOEs.

clean spectra from problematic NOEs (noise, artifacts), based on violation statistics. These NOEs can be printed for inspection, and their bounds can be re-set automatically, or they can be automatically excluded.

merge several independent data sets (different temperatures, solvents, homo- and heteronuclear).

calculate structures with X-PLOR, using either ab initio simulated annealing or a hybrid distance-geometry simulated annealing approach. Only the former allows full use of ambiguous NOEs in all stages of the calculation. Prochiral groups are treated with a floating chirality approach that is fully integrated into the treatment and assignment of ambiguous NOEs.

analyse and select structures for a next iteration of assignment/structure calculation.

Each data set is interpreted (calibrated, assigned, and cleaned) separately, before the different sets are merged to give one distance restraint list. Different types of data sets can be mixed; the operations performed by ARIA are determined by simple flags. Thus, one can use a list of peaks from an automatic peak picker, where calibration, assignment, and cleaning have to be performed. A single file (called project.xplor) needs to be edited to set all flags and parameters for the spectra assignment and the calculation.

Overview

1. generate a series of random structures and treat these as the basis of all operations (calibration, assignment, violation analysis and merging);

2. calculate a set of structures based on manually assigned, `hard' restraints;

3. generate starting structures from sequence homology.

Organization of the project

data This directory contains a subdirectory for the sequence related data, and for each dataset used (NOEs, and other data such as dihedral angles, etc.). It also contains a list of the names of all assigned spectra to be merged (spectra_ass.list), which is created at present by hand, for example:

sequence This subdirectory contains all sequence-related information such as the psf file, the template file for the generation of random starting structures, and a list of prochiral groups for which floating chirality assignment is performed. It also contains the file read_data_xplor, which needs to specify all data files read in for the structure calculation.

noe1 Each of the noe subdirectories contains the original peak file, the chemical shift list valid for this spectrum, and the converted restrained lists readable by X-PLOR. The allowed frequency ranges are determined during conversion, and cannot be changed later.

noe2 A different peak or restraint list.

hbonds This directory contains a list of hydrogen bonding restraints, which can be assigned or ambiguous.

torsion Standard torsion angle restraints.

assign Contains input files and X-PLOR include files to perform the partial assignment, cleaning and merging.

calib.inp uses the include files calibrate.xplor to perform calibration and errset.xplor is used to set error bounds. It performs a violation statistic to reset bounds of or remove certain restraints, and assigns ambiguous NOEs and writes the resulting restraints in the current iteration directory.

merge.inp takes the resulting restraints files of all NOE spectra and merges them to remove multiple entries arising from multiple picking of the connectivity.

protocol The read-only files of the simulated annealing protocol.

toppar The energy parameters used for the structure determination.

The subdirectories analysis, protocol and assign are (mostly) read-only and usually need not be modified. toppar also should not have to be modified, but it seems advantageous to keep exactly the parameters used with the project for documentation.

Start of a project

Floating chirality: Define the groups for which floating assignment is to be performed in the file setup_swap.xplor in data/sequence.

NOE data: The NOE data is organized by using a separate directory for each data set.

1. Create a separate directory for each NOE list to be used in the structure calculation.

2. Copy the original peak list and chemical shift information into this directory.

3. Edit an appropriate script from the assign directory to apply appropriate _ppm values in the two (three) frequency dimensions, and translate the peak files and chemical shift lists to the X-PLOR readable format.

spectra_ass.list: Create a file in directory data that contains a list of all file names of assigned spectra (the $restraintfile variable in project.xplor, prefixed with "@@NEWIT".

initialize.xplor

project.xplor

The second section contains file names of the structure file, the template file, and the file root for all coordinate output files.

The third section contains parameters for each dataset used in the refinement.

The fourth section contains parameters for each iteration (any of the above parameters can be put here if they are to be varied from iteration to iteration).

The last section contains some modifications of parameters defined above in order to specify correct directories (via environment variables).

X-PLOR/ARIA commands

<ANALysis-restraints-statement>:==

HELP

EQUIvalent <group> { <equivalent-statement> } END marks protons with unresolved chemical shifts as a single spin. The distance to an unresolved group is calculated as <r^-av>1/av where the exponent av is set with the AVERage command. The main use of this command is for methyl groups and aromatic rings.

<equivalent-statement>:==

1-2 treats each methylene group as an unresolved group of protons. A methylene group is recognized by X-PLOR as three hydrogens bound to a heavy atom. At present not working.

METHyl treats each methyl group as an unresolved group of protons. A methyl group is recognized by X-PLOR as three hydrogens bound to a heavy atom. At present not working.

NONE deletes all previously defined unresolved groups.

SELECT=<selection> treats the selected atoms as an unresolved group of protons.

SINGle reads in distances from the main coordinate set and overwrites any existing data.

ACCumulate adds distances from the main coordinate set, calculating a <r^-6>-^1/6 average.

AVERage adds distances from the main coordinate set, calculating an arithmetic average.

MINImum adds distances from the main coordinate set, calculating the minimum value.

MODE=INTRA-residue | ALL INTRA will count as ambiguous only NOEs that involve more that one residue pair; ALL all ambiguities (default: ALL).

LEVEl=<real> Assignment criterion. Possibilities that have an estimated contribution to the peak of less than LEVEl will be removed.

CUTOff=<real> Assignment criterion. Possibilities that correspond to distances larger than CUTOff will be removed.

FRQUpdate <class> Updates frequencies (in RMSD array) to the weighted average of the NOE peak positions.

CHECk <class> Removes duplicate entries on restraint list. All restraints are compared to each other. If two restraints have the same assignment, the one with narrower bounds will be kept.

PURGe <class> <class> Remove an unambiguously assigned assignment possibility in the first class from all restraints in the second class.

TRANsfer not implemented.

MINN=<integer> Minimum number of possibilities to be written. If a restraint has less than the specified number of possibilities, it will not be written.

MAXN=<integer> Maximum number of possibilities to be written. If a restraint has more than the specified number of possibilities, it will not be written.

FROM=<selection>...TO=<selection> Write only restraints involving specified atom pairs.

LIST<class> Write out restraints as a human readable list in atom order.

SORT<class> Write out restraints as human readable list in order of estimated contribution.

RESTraints <class> Write out X-PLOR restraint list using standard atom selection to specify ambiguities.

OR-Restraints <class> Write out X-PLOR restraint list specifying ambiguities with OR between double selections.

RESET Reinitialize database.

HELP

RESEt Reinitialize database.

THREshold <real> Count the number of times a restraint is violated in an ensemble of structures.

EXCLude <class><real> Exclude all restraints in a class that are violated in more than the fraction of structures specified.

LIST <class><class> List all restraints in a class that are violated in more than the fraction of structures specified.

SET <class> <real> <real> <real> Set limits for all restraints in a class that are violated in more than the fraction of structures specified to the specified lower and upper bound.

HELP

RESEt Reinitialize database.

MODE (ANISotrop | ISOTrop) By default, a calibration factor is calculated for each pair of proton groups separately. The ISOtropic mode re-averages these calibration factors such that C_IJ=(C_II.C_IJ)^1/2. This is at the moment somewhat unreliable.

CUTOff <real> Exclude all distances in trial structures that exceed cutoff.

VECTor <sele> <sele> ( AUTO | DRED | VREF ). If AUTO is specified, the calibration distances are taken as averages over all interproton distances in the trial structures for which there is an NOE. Explicit reference distances and volumes can be specified with DREF and VREF commands. The missing terms in the matrix of calibration factors are then automatically calculated using the isotropic approximation.

DREFerence <real> Specify an explicit reference distance for the last specified pair of proton groups.

VREFerence <real> Specify an explicit reference volume for the last specified proton groups.

EXPOnent <real> Specify an explicit reference volume for all protons.

J-coupling and proton and ¹³C13 carbon chemical shift NMR refinement

The syntax required for direct coupling refinement is a set of statements of the form COUP E<coupling statement>L END

with

<COUPLINGS-statement>:==

ASSIgn<slctn><slctn><slctn><slctn>[<slctn><slctn><slctn><slctn>]

<real><real>[<real><real>]

E* atom i , atom j , atom k, atom l, J-obs, J-err *L.

The optional second set of slctns and obss is for degenerate classes

CLASs <name>

Starts a new class. Applies to all ASSIgn, TYPE, FORCe and FLAT entries until another CLASS entry is issued.

COEFFicients <real><real><real><real>

Set the Karplus curve coefficients A,B,C and the Karplus curve phase P (in that order) for this class.

CV=<integer>

Select partition number for cross-validation.

DEGEneracy <1 | 2>

Set the number of couplings to be averaged for each observation in this class (Edefault = 1L).

FORCe <real><real>

Force constant(s) for all assignments in the current class (Edefault = 50, 10L).

NREStraints <integer>

Number of slots for J coupling.

PARTition=<integer>

Number of partitions for complete cross-validation.

POTential <SQUAre | HARMonic >

Use J-err or not.

PRINt THREshold <real> <ALL | CLASs <name>>

Prints J-coupling violations greater than the specified value (in Hz).

RESet

Erases the J coupling assignment table.

The syntax required for direct coupling refinement is a set of statements of the form CARB {<chemical-shift-statement>} END

with

<CHEMSHIFT-statement>:==

ASSIgn <sel><sel><sel><sel><sel><real><real><real><real>

{* Ci-1, Ni, Cai, Ci, Ni+1, Ca-obs, Cb-obs (in ppm) *}.

CLASsification <name>

Starts a new class. Applies to all ASSIgn and FORCe entries until another CLASS entry is issued.

EXPEctation <integer> <integer> <real> <real> <real> <real>

Sets (psi-pos, psi-pos, CAvalue, CAerr, CBvalue, CBerr ) for expectation values.

FORCeconstant <real>

Force constant for all assignments in the current class (default = 50).

NREStraints <integer>

Number of slots for chemical shift restraints to allocate in memory (default = 200).

PHIStep <real>

Number of steps in the phi dimension of the expectation array.

PSIStep <real>

Number of steps in the psi dimension of the expectation array.

POTEntial <SQUAre | HARMonic>

Use shift errors or not.

PRINt THREshold <real>

Prints secondary shift violations of either Ca and Cb greater than the specified value (in ppm).

RCOIl <sel> <real> <real>

Set the random coil a and b ¹³C shifts for the selected atoms.

RESEt

Erases the chemical shift assignment table.

ZERO

Zero out the expectation value arrays.

The syntax required for direct ¹H chemical shift refinement is a set of statements of the form PROTONSHIFTS {proton-shift-statement} END

with

<PROTONSHIFTS-statement>:==

OBSErved <sel> [<sel>] <real> [<real>]

Set the observed proton shift or shifts for the protons in the selection(s) to the real value(s).

RCOIl <sel> <real>

Set the random coil proton shift for all the protons in the selection to the real value(s).

ANISotropy <sel> <sel> <sel> <CO|CN> <isCOOH> <SC | BB>

Select CA, CO,O atoms (one in each selection), say if it is a CO or CN bond, if it is part of a carboxyl group (which are averaged together), and whether it is a backbone or sidechain CO.

AMIDes <sel>

Select all the backbone amide protons.

CARBons <sel>

Select all the BB carbonyl carbon atoms.

NITRogens <sel>

Select all the BB nitrogen atoms.

OXYGens <sel>

Select all the BB oxygen atoms.

RINGatoms <PHE|TYR|HIS|TRP5|TRP6|ADE6|ADE5|GUA6|GUA5|THY| CYT|URA> <sel> <sel> <sel> <sel> <sel> [<sel>]

Specify how many atoms are in the ring, and select each one in turn. The last selection is only for six-member rings.

ALPHasandamides <sel>

Select both alpha and amide protons.

CLASsification <name>

Starts a new class. Applies to all ASSIgn and FORCe entries until another CLASS entry is issued.

ERROr <real>

Specify the error to be used by the current class in square well calculations (default = 0.3).

DEGEneracy <1 | 2>

Set the number of shifts to be averaged for each observation in this class (default = 1).

FORCeconstant <real> [<real>]

Specify the force constant and Constantine force constant for all assignments in the current class (defaults = 0.1).

POTEntial <SQUAre | HARMonic>

Specify whether to use se shift errors or not.

PRINt THREshold <real> <ALL | CLASs <name>> <RMSD|NORMsd>

Print proton shift violations greater than the specified value (in ppm) for all classes or for the specified class. May also use to prints the results to the RMSD array

RESEt

Erases the proton shift assignment table.

The technology of the calculation of NOE intensities and the gradients with respect to the atomic coordinates have improved over the last few years. The method that is implemented in X-PLOR is based on the ideas of matrix doubling and Gaussian quadrature (Yip 1993). This present implementation is significantly faster than the earlier implementation based on Yip and Case (1989). To date, accelerations of one to two orders of magnitudes have been obtained, depending on the system studied.

Implementation

MODE=<NONE,GRADient,ALL,MATD>

applies no cutoff, cutoff only for the gradient calculation, cutoff for both intensity and gradient calculation, and cutoff employed for both intensity and gradient calculation but using the new matrix doubling method (default: none).

When MATD is selected, one must specify the value of the cutoff by the parameter VALUe,. The RELAtive_value parameter is inoperative. A value of 6-8 is recommended for VALUe when MATD is chosen.