Commands for controlling how SOLVE works in automated mode
This is a list of the most commonly-used keywords for SOLVE that apply to SOLVE operation in automated mode. Also see the list of all SOLVE keywords and the list of all commands .
These keywords are usually in your solve.setup file:
SYMFILE xxxxx symmetry file for this space group
CELL a b c alpha beta gamma
RESOLUTION dmin dmax
FFTGRID xs xe xtotal ys ye ytotal zs ze ztotal grid for FFT calculations
PATTGRID xs xe xtotal ys ye ytotal zs ze ztotal grid for Patterson
EZDGRID xs xe ys ye zs ze grid for NEWEZD map
These keywords are usually in your main command file:
VERBOSE write out a lot of output to logfile
RAWMADFILE xxx.int File xxx.int will be read in as data for
the current mad wavelength
RAWNATIVEFILE xxx.int File &xxx.int& will be read in as data for
the native
RAWDERIVFILE xxx.int File xxx.int will be read in as data for
the current derivative
READ_INTENSITITES (default) The raw data files contain intensity measurements
READ_AMPLITUDES The raw data files contain amplitudes (F) not intensities (I)
(This is valid only with READFORMATTED)
PREMERGED The data in all RAWMADFILEs have H K L and 4 other columns:
I+/F+, sigma, I-/F-, sigma
UNMERGED The data in all RAWMADFILEs have H K L and 2 other columns:
I/F, sigma
READDENZO All datafiles are written by Scalepack. For unmerged data
they will be read with the formatting:(6i4,i6,2i2,i3,2f8.0) and
nsym*2+1 lines are skipped at the top of the file. For
merged data the formatting is: (3i4,4f8.0) and
3 lines are skipped at the top of the file.
READFORMATTED All datafiles will be read with "*" formatting and
contain H K L I/F sigma or H K L I+/F+ sigma I-/F- sigma
READTREK The datafiles were written by d*trek and contain columns
with intensities
NSKIP n Skip exactly n lines at the top of each data file
NSKIP 0 Do not skip any lines at the top of each data file
NSKIP -1 Skip 0 lines at the top of each data file
unless the keywords READDENZO and PREMERGED
are set in which case the default number of lines
are skipped (see above)
RATMIN xx Minumum ratio of F/sig to read in data for a
reflection at all is xx [default=2.0]. This is
useful for eliminating weak data.
FPFM_ONLY Toss all acentric reflections where either F+
or F- is missing [this is the default for MAD data]
FP_OR_FM Use F+ or F- as an estimate of Fbar if F+ and
F- are not both present. This is useful if
your data is not that complete. It is much
better to obtain complete data however.
OVERALLSCALE Do not do local scaling; just an overall
scale factor for F+, F- at each wavelength.
Use this if you already have scaled the data
and you don't want any more scaling done.
SWAP_ANO Swap H K L -> -H -K -L as data are read in to
SOLVE in scale_native, scale_derivative, and
scale_mad. This is to correct for a detector or indexing
that swapped F+ for F-
The atom types recognized by SOLVE are: S, Fe, Cu, Se, Ru, Rh, Pd,
Te, I, Xe, Sm, Eu, Gd, Tb, Ho, Yb, Hf, Ta, W, Re, Os, Ir, Pt,
Au, Hg, Pb, Th, and U
newatomtype xxxx define scattering properties of atom xxxx not recognized
by SOLVE
HINT: to get the aval, bval, cval values from Int tables
for atoms recognized by SOLVE, type mad_atom [atomname] and
SOLVE will list them for you.
aval a1 a2 a3 a4 4 real numbers (a1,a2,a3,a4) from International Tables for
the most recently defined newatomtype
bval b1 b2 b3 b4 b values for newatomtype
cval c c value for newatomtype
fprimv xx f' value for newatomtype
fprprv xx f" value for newatomtype
mad_atom xxxx name of the anomalously scattering atom is xxxx. If SOLVE does
not recognize this atom, first input its scattering parameters
with newatomtype
fprimv_mad f' value for anomalously scattering atom
at this wavelength (must be input after each wavelength)
fprprv_mad f" value for this wavelength
FIXSCATTFACTORS Fix scattering factors at their input values.
This is a good idea if you have a reasonable
idea of the f' and f" values. [this is the
default]
REFSCATTFACTORS refine scattering factors f' and f". If you
refine them, be sure to look at their new
values at the end of the routine MADMRG and
verify that they are reasonable.
derivative n begin input of information for derivative/wavelength n
This command is used to start entering information on a
derivative. If you want to modify something after you've
gone on to another derivative then you need to use the
command GOTODERIV
lambda n identical to derivative n
nsolsite_deriv Maximum number of sites for this derivative
(overrides nsolsite)
cutoff_deriv resolution limits for this derivative/wavelength only
INPHASE include this wavelength/derivative in phasing.
NOINPHASE do not this include this derivative/wavelength in phasing
INANO include anomalous differences for this wavelength/deriv
noanorefine use anomalous differences in phasing but not
refinement for this derivative.
(this is usually the best option for MIR
unless your anomalous differences are really
big, as from a synchrotron MIR dataset at an
absorption edge). Note: you still have to specify
for each derivative "inano" to include anomalous
differences for that derivative.
anorefine For this derivative with "inano" specified, use
anomalous differences in both refinement and phasing.
This is best for MAD data. (This is the default also).
Applies to current derivative/wavelength
ATOMNAME XXX Start reading in information about a new site with atomname
"XXX" at the current wavelength/deriv.
OCCUPANCY xx occupancy of current atom
BVALUE xx B-value of current atom
XYZ xx yy zz fractional coordinates of current atom
REFINEALL refine x,y,z,occ and b for this atom
REFINENONE don't refine anything for this atom
NRES n # of residues in asymmetric unit [default=100]
NANOMALOUS n # of anomalously scattering atoms in asymmetric unit.
Used to estimate how big the Fa values might be.
Also used to set max # of heavy atoms if nsolsite is not set
nobayes do not use Bayesian correlated phasing in SOLVE.
ntopfour xx Number of Fourier peaks to pick from a map
ntopderiv xx Number of Fourier peaks to be tested for
inclusion in the model
nsolsite xx Maximum number of sites in a derivative unless overridden by
nsolsite_deriv
nseedtest xx Number of seeds per derivative to try (before
sorting)
nseedsolve xx Number of seeds (total) to try after
sorting them
ntopsolve xx Number of solutions to print out at the end
and number of solutions to keep track of at
any one time
addsolve Add on to solution that is input
[default=off]
checksolve Compare all solutions to input solution
[default=off]
analyze_solve Analyze input solution without doing anything
else [default=off]
[no]delete do [not] check out all solutions by testing
all one-site deletions [default=delete]
[no]inverse do [not] check out all solutions by testing
their inverses (does not apply if a solution
is centrosymmetric or if anomalous differences
are not used). [default=inverse]
SCORING_TABLE (8 values) Scoring table (usually generated by SOLVE) consisting
of mean and standard deviation of scores for trial
solutions for Pattersons, Cross-fouriers, Native Fourier
maps, and mean figure of merit. This keyword is useful
when you are running SOLVE after modifying the script
file it writes out at the end.
QUICK once a plausible solution is found, don't keep looking,
just add on sites to it and check it at the end. [default]
THOROUGH keep looking anyways until a limit set by ntopsolve,
nseedsolve, etc is reached.
NTOL_SITE a site within ntol_site grid units of an existing site is
considered to be a duplicate and is ignored. [default=8]
NTOL_SOLN a heavy-atom solution for which every site matches another
solution within ntol_soln grid units is considered to be
a duplicate and is ignored. [default=2]
ACCEPTANCE xx the weighting function for scoring patterson and free-
difference fourier peak heights is adjusted so that a new
site with height relative to the previous average height
of ACCEPTANCE or higher will generally give a solution
with a higher score than the solution without this site.
[default =0.2]