REFMAC (CCP4: Supported Program)
User's manual for the program refmac_5.0.36
Keyworded input - Restraints keywords
Anything input on a line after "!" or
"#" is ignored and lines can be continued by using a minus (-) sign.
The program only checks the first 4 characters of each keyword. The order
of the cards is not important except that an END card must be last. Some
keywords have various subsidiary keywords. The available keywords in this section
- Restraints on bond angles
- Restraints on B values
- Restraints on chiral volumes
- Restraints on bond distances
- Restraints against excessive shifts
- Controls making restraints and checking coordinates against dictionary
- Restraints on non-crystallographic symmetry
- Restraints on planarity
- Rigid bond restraints on the anisotropic B values of bonded atoms
- Sphericity restraints on the anisotropic B values
- Restraints on the torsion angles
- Restraints on VDW repulsions
These keywords are used to create restraints and to set the
weighting of the restraints.
MAKE_restraints [HYDR Y|N|A] [HOUT Y|N] [CHECk Y|N]
[BUIL Y|N] [FORM F|U] [PEPT Y|N] [LINK Y|N] [SUGAr Y|N|D]\
[CONNectivity Y|N|D] [SYMM Y|N] [CISP Y|N] [SS Y|N|D] [CHAIn Y|N]
[NEWLigand Exit|Noexit] [VALUe COOR|ENER] [EXIT Y|N]
For full description including algorithms look MAKECIF"s documentation
This keyword controls the level of automatic restraint creation. REFMAC uses
predefined dictionary entries.
Any additional entry could be added using LIB_IN <file name> which contains a user's local ligand descriptions (see
How to make new ligand dictionary?).
On the basis of these files, the program makes intelligent suggestions on potential extra
restraints. If additional user input is needed, the program stops, outputting information
that should help the user to decide what to do next (e.g. a coarse picture of a
ligand as the program 'understands' it, for the user to help in setting up any extra
MAKE HYDR All
MAKE HOUT No
MAKE CHEC No
MAKE BUIL No
MAKE FORM Unformatted
MAKE PEPT No
MAKE LINK No
MAKE SUGA Yes
MAKE SS Yes
MAKE CISP Yes
MAKE SYMM No
MAKE CONN No
MAKE CHAIn Yes
MAKE NEWLigand Exit
MAKE VALUe ENERgy
MAKE EXIT No
- HYDR [ Yes | No | All ] (Default HYDRogens All)
- How to deal with the hydrogen atoms.
- Y - if hydrogens are present in the input file, use them,
- N - ignore hydrogens even if they are present in the input coordinate file,
- A - add all hydrogens in their riding positions and use them for geometry and structure
factor calculations. They also contribute to the gradient and second derivative matrix of
- For low resolution and early stages it is better to use MAKE HYDR N,
i.e. do not use hydrogens even if they are present in the input file.
- HOUT [ Yes | No ] (Default HOUT N)
- Whether to write hydrogens to the output coordinate file or not.
- Y - write hydrogens to the output file,
- N - do not write hydrogens to the output file.
- CHECk [ ALL | LIGAnd | NONE ] (Default CHECk LIGAnd)
- Whether to check amino acids in the input coordinate file against dictionary
- ALL - check all monomers against dictionary descriptions,
- LIGAnd - do not check amino acids, DNA/RNA and standard sugars. In general monomers which are part of chain will not be checked. It is safe option to use.
- NONE - do not check all monomers. In this case program relies on correctness of monomer and atom names in monomers. User has to be very careful in using this option.
- BUILd [ Yes | No ] (Default BUILd N)
- Whether to build absent atoms or not. For example, if you have
ASN in input coordinates and OD1 and ND2 are absent, the program can try to
find their position. Usually rebuilt atoms have occupancy 0.00, so they do not
- Y - build absent atoms,
- N - do not build absent atoms.
- FORM [ Formatted | Unformatted ] (Default FORMat Unformatted)
- Whether to use restraints file in Formatted or Unformatted
form. If you want to examine the list of restraints then you could use:
MAKE FORM F
In other cases unformatted form seems to be faster to handle.
MAKE EXIT Y
- PEPT [ Yes | No ] (Default PEPTide N)
- Check and/or use presence of L or D peptides.
- Y - check if peptide is L or D and use as they are,
- N - check if there are D peptides but use standard L peptide descriptions.
- LINK [ Yes | No | Define ] (Default LINK N)
- Check and/or use links between different residues that are not involved in standard peptide links
(i.e. connectivity that is not implied by the primary structure. Atoms involved in bonds
between HET groups, or between a HET group and standard residues, as well as reduced peptide bonds
can be described with this keyword).
- Y - check and, if specified by record LINK in the input coordinate file, use links
between residues. Links which have not been specified in the input coordinates but are found
by the program, will be added to the list of links specified in the input coordinate file.
- N - check links between residues and give information about them in the log file,
but do not use them UNLESS they are specified by record LINK in the input coordinate file.
- D - ignore links in the input coordinate file specified by record LINK. Check if current
coordinates suggest links, and ONLY use those; i.e. let the program define
what is a link and what is not.
In the early stages it is better to use LINK N. In later stages the user can use the
following keywords to check if there are any links between residues which should be taken
MAKE FORM F
MAKE LINK D
MAKE EXIT Y
After creation of restraints and checking all links, the program will stop. Then the user
needs to examine the .log file, restraint file and the new library suggested by the program,
to see if there is anything worthy of more attention.
- Examples of a LINK record in a coordinate file:
1 2 3 4 5 6 7
LINK O1 DDA 1 C3 DDL 2
LINK MN MN 391 OE2 GLU 217 2565
- SUGAr [ Yes | No | Define ] (Default SUGAr Y)
- Check and/or use links between sugars and sugar-protein links. These links can be specified
through record LINK in coordinate files.
- Y - check and, if specified by record LINK in the input coordinate file, use links between
sugars and sugar-protein links. If sugar links are present in the input coordinate file
they will have high priority.
- N - check links between sugars and sugar-protein links and give information about them
in the log file, but do not use them UNLESS they are specified by record LINK in the input
- D - ignore sugar links in the input coordinate file specified by record LINK. Check if
current coordinates suggest links, and ONLY use those; i.e. let the program define
what is a suitable sugar link and what is not.
- SSbridge [ Yes | No ] (Default SS Y)
- Check and/or use disulphide bridges automatically.
- Y - the program will find disulphide bridges and use restraints
- N - the program will check presence of disulphide bridges but ONLY use disulphide bridges
if they are specified in the input coordinate file through record SSBOND.
- CISP [ Yes | No ] (Default CISPeptide Y)
- Check and/or use cis peptides automatically.
- Y - find and use cis peptides automatically,
- N - check but do not use cis peptides UNLESS they have been given specifically
in the input coordinate file, through record CISPEP.
- SYMM [ Yes | No ] (Default SYMM N)
- Check and/or use links between symmetry related atoms.
- Y - check the existence of links between symmetry related atoms and use if described
- N - do not use links between symmetry related atoms.
- CONN [ Yes | No | Define ] (Default CONNectivity N)
- Check connectivity between consecutive residues (e.g. standard peptide links and connectivity
within a HET group).
- Y - check and, if specified by record CONECT in the input coordinate file, use connectivity in
polypeptides and DNA/RNA. Connectivity which is not specified in the input coordinates but is found
by the program, will be added to the list of connectivity specified in the input coordinate file.
- N - check connectivity between consecutive residues of polypeptide chain and DNA/RNA, but do not use
them UNLESS they are specified by record CONECT in the input coordinate file.
If consecutive residue numbers differ by more than two, it is assumed there is no link between them.
- D - ignore connectivity in the input coordinate file specified by record CONECT. Check if coordinates
suggest connectivity, and ONLY use those; i.e. let the program define what is
connected and what is not.
- CHAIn [ Yes | No ] (Default CHAIn N)
- Check and/or use chain definition automatically.
- Y - check chain definitions for peptide/DNA/RNA/sugar, cyclic peptide etc.,
and use if they are suggested by the coordinate file.
- N - check but do not use chain definitions.
- NEWLigand [ Exit | Noexit ] (Default NEWLigand Exit)
- If the program encounters new ligand, it will stop and tell what to do next.
MAKE NEWLigand Noexit
has been specified then the program will continue to work relying on
the dictionary description it has created using coordinates and/or minimum
description of this ligand. At the moment it is highly recommended to check
entries created by the program and make sure that these correspond to
the ligands the user wants.
- VALUe [ COORdinates | ENERgy ] (Default VALUe ENERgy)
- It tells to the program where to take the ideal values. If 'ENERgy' has
specified then ideal values will be taken from the energetic library. If
'COORdinates' has been specified then the program will take ideal values from
the coordinates. If you want to take ideal values from the coordinates then
make sure that coordinates have been derived using good energetic minimisers or
they have been refined against very high resolution data.
- EXIT [ Yes | No ] (Default EXIT N)
- It tells the program to exit or not after creating restraints. It is
useful when a new entry is added. Then the user could use following keywords:
Then look at the restraints file and check validity and values of the restraints.
MAKE FORM F
MAKE EXIT Y
<WDSKAL> is a multiplier used in calculating the weights for the
distance restraints. The weights are of the form: Weight = (WDSKAL/ SIGD)2.
Sigmas for bond distances come from the dictionary file.
<angle_scale> is multiplier used in calculating weights for the
bond angle restraints. The weights have the form: Weight = (angle_scale/SIG_angle)2.
SIG_angle comes from the dictionary file.
<WPSKAL> is a multiplier for the distance restraints defining
planes. The weight used for the planes is (WPSKAL / SIGPlane)2.
SIGPlane comes from the dictionary file.
[Default 1.0 0.15]
<WCSKAL> is used in weighting Chiral groups restraints. The weight
used is (WCSKAL /SIGChiral)2.
SIGChiral comes from the dictionary file.
TEMPerature | BFAC <wbskal> <sigb1>
<sigb2> <sigb3> <sigb4>
TEMPerature | BFAC SET <B_value>
BFAC 1.0 1.5 2.0 3.0 4.5
BFACtor SET 20.0
[Default 1.0 2.0 4.0 4.0 6.0 ]
<WBSKAL> is used in calculating the weight for the temperature
factor restraint parameters based on the types of bonding in which the
atoms are involved. Weight calculated for B_value restraints is
- sigma for bonded atoms of the main chain
- for bonded atoms of the side chain
- for angle related atoms of main chain
- for angle related atoms of side chain
SET <B_value> - tells the program to set all B values to <B_value>
<sigsph> is used to restrain atomic anisotropic tensor to be
spherical. Used weight is calculated as 1.0/<sigsph>**2.
[Default 2.0 ]
<sigrbond> is used to make minimum of projections of anisotropic tensors
along bond for the bonded atoms. Used weight is calculated as 1.0/<sigrbond>**2.
NCSR [NCHAIN <nchains>]
[NSPAns <nspans> <residue_first_1> <residue_last_1> <ncode_1>.....
<residue_first_nspans> <residue_last_nspans> <ncode_nspans>]
To define non-crystallographic symmetry. Previously part of PROTIN. Now
moved to REFMAC.
NCSRestraints NCHAins 5 CHAIns A B C D E NSPANS 2 1 100 1 101 150 2
NCSR <wsscal> <sgsp1> <sigsp2>
<sigsp3> <sigsb1> <sigsb2> <sigsb3>
Parameters for weighting of the contribution of the ncs restraints.
For example (they are default values):
NCSR 1.0 0.05 0.5 5.0 0.5 2.0 10.0
This keyword controls restraints on non-crystallographically related atoms.
If the first value after NCSR is NCHAin, then the first option, i.e. definition of
NCS restraints, is assumed. In this case, the following subkeywords are available:
- NCHAIns <nchains>
- Number of chains involved in this NCS restraint
- CHAIns <chain_id_1>.....<chain_id_nchains>
- Identifier of the chains involved in this NCS restraint.
Number of chain identifiers should be equal to <nchains>.
- NSPAns <nspans> <residue_first_1> <residue_last_1> <ncode_1>.....<residue_first_nspans> <residue_last_nspans> <ncode_nspans>
- Number of pieces of chains which should be treated with different weighting
- <nspans> is number of spans,
- <residue_first_i><residue_last_i> are first, last residue numbers involved in i'th span, with
- <ncode_i> is weighting to be used for i'th span, with i=1,nspans.
- Codes for restraints are as follows:
ncode Main_Chain Side_Chain
1 tight restraint tight restraint
2 tight restraint medium restraint
3 tight restraint loose restraint
4 medium restraint medium restraint
5 medium restraint loose restraint
6 loose restraint loose restraint
If NCSR is followed only by numbers, these values are used for weighting of
contribution of NCS restraints.
[Default 1.0 0.05 0.5 5.0 0.5 2.0 10.0]
<wsscal> is used in weighting restraints involving non-crystallographic
symmetry. The weight is given by (<wsscal> / <sigs>)**2.
<sigsp1>, <sigsp2>, <sigsp3> are values associated
with non-crystallographic positional restraints and are for tight, medium
and loose restraints respectively.
<sigsb1>, <sigsb2>, <sigsb3> are values associated
with non-crystallographic thermal restraints and are for tight, medium
and loose restraints respectively.
<wtskal> is a multiplier used in calculating the weights for the
torsional angle restraints. The weight is of the form Weight = (<wtskal>
/ <SIGT>)**2, SIGT comes from dictionary files.
VANDerwaals (or VDWR or NONBonding) <wvskal>
VANDerwaal (or VDWR or NONBonding) [OVERall <wcskal>] [SIGMA VDW <vsigma > |
HBONd <hsigma> | METAl <msigma> | TORSion <tsigma> DUMMy <dsigma> ]
[INCRement TORSion <tincrement> | ADHB <adincrement> | AHHB <ahincrement> DUMMy <dincrement>]
NONBonding OVERall 1.0
NONBonding SIGMa VDW 0.3
NONBonding SIGMa HBOND 0.5
NONBonding SIGMa METAl 0.5
NONBonding SIGMa TORSion 0.5
NONBonding SIGMa DUMMy 0.3
NONBonding INCRement TORSion -0.75
NONBonding INCRement ADHB -0.2
NONBonding INCRement AHHB 0.2
NONBonding INCRement DUMMy -0.7 ]
This keywords controls parameters and weights of restraints on nonbonding
interactions. REFMAC deals only with the repulsive part of nonbonding interactions, i.e.
if separation of the atoms is larger than "ideal" then they
are not considered to be interacting atoms.
The program excludes all atom pairs related by one covalent bond, angle, or if
they are in the same plane. Atoms related by one torsion angle are treated
- OVERall <wvskal>
- Controls overall weights on all nonbonding interactions.
- SIGMa VDW <vsigma> | HBONd <hsigma> | METAl <msigma> | TORSion <tsigma> DUMMy <dsigma>
- Controls sigmas for each type of nonbonding interactions.
- VDW <vsigma>
- sigma of atom pair involved in vdw repulsion
- HBONd <hsigma>
- sigma of atom pair potentially involved in hydrogen bonding interactions
- METAl <msigma>
- sigma of atom pair potentially involved in metal-ion interactions
- TORSion <tsigma>
- sigma of atom pair involved in vdw interactions and related by one torsion angle.
- DUMMy <dsigma>
- sigma of atom pair involved dummy and nondummy atoms.
- INCRement TORSion <tincrement> | ADHB <adincrement> | AHHB <ahincrement>
- Provides increment factors for different types of nonbonding interactions.
- TORSion <tincrement>
- If atoms are related by one torsion angle and
they are closer than vdw1 + vdw2 + tincrement, then they are
considered to be interacting atoms. Here vdw1 and vdw1 are
vdw radii of the interacting atoms.
- ADHB <adincrement>
- If atoms can make hydrogen bond, i.e. one of them
is a potential acceptor and the other is a potential donor and they are closer than
vdw1 + vdw2 + adincrement, then they are considered to be
interacting atoms. Here vdw1 and vdw2 are vdw radii of
donor and acceptor.
- AHHB <ahincrement>
- If one of atoms is an acceptor and the other one a hydrogen
from a donor and they are closer than vdw1 + ahincrement, then they are considered
to be interacting atoms. Here vdw1 is vdw radius of the
- DUMMy <dincrement>
- If on of atoms is dummy and another is
nondummy and they are closer than vdw1 + vdw2 + dincrement, then they are considered as interacting atoms. Here vdw1 and vdw2 are vdw radii of atoms involved in the interaction.
HOLD <PBEL> <BDEL> <QDEL>
[Default 0.3 3.0 0.2]
Restraint to current values <PDEL> is a restraint on the magnitude of positional
shifts. This goes into the matrix as (a/<PDEL>)**2, (b/<PDEL>)**2,
(c/<PDEL>)**2 where a, b, c are the unit cell parameters.
<BDEL> is a shifts magnitude restraint on individual thermal parameters
and goes into the matrix as (1/<BDEL>)**2. It is not used if ITEMP = 0.
<QDEL> is the shifts magnitude restraint on variable occupancy
factors (not used if NOCC=0). This goes into the matrix as (1/<QDEL>)**2.
[See also keyword MONItor]