Go to the previous, next section.
The commands described in this node are used to construct and manipulate the PSF, the central data structure in CONGEN. The PSF is comprised of lists giving every bond, bond angle, torsion angle, and improper torsion angle as well as information needed to generate the hydrogen bonds and the non-bonded list. It is essential for the calculation of the energy of the system. A separate data structure deals with symmetric images of the atoms (see section Symmetry and Molecular Images).
There exists one other command for manipulating the PSF, the SPLICE command, see section SPLICE -- Change the Sequence of the PSF.. The SPLICE command can change the sequence of PSF and shuffle the coordinates so that the CONGEN command, see section Conformational Search, which can be used to find conformations for the new amino acids.
There is an order with which commands to generate and manipulate the PSF must be given. First, segments in the PSF must be generated one at a time. Prior to generating any segments, one must first have read a residue topology file, see section Residue Topology Files. To generate one segment, one must first read in a sequence using the READ command, see section Specifying a Sequence of Residues for a Segment. Then, the GENERATE command must be given. RTF's may be changed as needed; however, the SPLICE command will only work correctly when only one RTF is used to build the entire structure.
Once a segment is generated, it may be manipulated. It may be edited using the EDIT command. Cystine bridges may be added using the DISULFIDE command. A histidine heme crosslink may be added using the PATCH HEME command.
GENErate [segid] [NBXMod int] [rtf-type] [CYCLic]
[[NO]ANGLes] [NOTORSions ] [[NO]DONOrs] [[NO]ACCEptors]
[TORSions {ALL}]
[ {ONE}]
{ PROT }
{ HPRO }
{ ALLH }
rtf-type ::= { DNA }
{ A94N }
{ A94P }
{ AM94 }
Using the sequence of residues specified in the last READ SEQUENCE command and the information stored in the residue topology file, this command generates the next segment in the PSF. Each segment contains a list of all the bonds, angles, dihedral angles, and improper torsions needed to calculate the energy. It also assigns charges to all the atoms, sets up the nonbonded exclusions list, and specifies hydrogen bond donors and acceptors. If a special type of segment has been specified in the READ SEQUENCE command or by the rtf-type option, modifications for structural features not contained in the residue topology file, for example terminal group modifications and proline modifications, are made automatically. The CYCLIC option controls whether a cyclic structure is built. Cyclic structures are made by omitting any terminal residues, and wrapping references to atoms beyond each end of the segment back to the other end.
The processing of terminal groups varies depending on the rtf-type setting. If a CONGEN topology file is read; rtf-type equals PROT, HPRO, ALLH, or DNA; then extra residues, like NTER and CTER, are added to the sequence. This has the undesirable side-effect of adding extra residues into your sequence, and confusing the residue numbering.
If the AMBER 94 potential is used, a different scheme is used. Here, the topology file, section AMBER94RTF, contains special terminal residues, which have different atoms and charges. The GENERATE command will translate the terminal residues to the names in the topology file, generate the segment, and then translate the names back. The following example table illustrates the naming conventions used for alanine.
The A94P keyword specifies that an AMBER 94 protein sequence is to be generated, and the A94N keyword specifies a nucleic acid sequence is to be generated.
The GENERATE command is capable of automatically generating some of the information needed to compute the energies from other sources within the PSF. In the case of the AMBER potential, CONGEN sets these options on by default. In addition, the value of these switches is saved in the PSF (in the NICTOT array), so that they can be used by the SPLICE command, see section SPLICE -- Change the Sequence of the PSF.. The automatic generation options have the following interpretations:
The actual generation process proceeds in four phases. First, all the atoms specified by the residues are added to the PSF. Next, all the terms are added, and all linkage references can be correctly handled, see section Linkage Atom Naming. Next, the automatic generation operations are performed. Finally, the patches are performed.
Some pairs of atoms are excluded from the nbond exclusion lists because their interactions are described by other terms in the hamiltonian. By default directly bonded atoms and the 1-3 atoms of an angle are excluded from the nonbond calculation. In addition the diagonal interactions of the six membered rings in tyrosine and phenylalanine and ring atom interactions in tryptophan are excluded in the current topology files. Hydrogen bonds, and dihedral 1-4 interactions are not excluded (note that other workers may differ from us on one or both of these points).
The list of nonbonded exclusions is generated in two steps. First a
preliminary list is made at generation by GENIC using any
information that may be present in the topology file (as for example
might be diagonal interactions in rings). The second step is an
automatic compilation of all the bond and angle interactions, followed
by a sorting of the list, performed in MAKINB. The list is
stored in the linked list pair IBLO/INB, where IBLO(i)
points to the last exclusion in INB to atom i. If the
list is modified after MAKINB, then either MAKINB should
be called again to resort the list, or care must be taken to see that
the INB list is ascending with all INB entries having
higher atom numbers than i and that all atoms have at least one
INB entry.
MAKINB is called by default after any operation which changes
internal coordinates such as generate, patch, edit, or splice.
The default list can be modified in three ways. First, interactions that are to be excluded can be placed in the topology file. Second, the NBXMOD option can be specified as a qualifier to any of the commands which change internal coordinates. Its values and actions are:
MAKINB
will correct this automatically.
Note that the 1-3 and 1-4 interactions are determined from examination of the bond list regardless of any torsions or improper torsions which are defined.
Negative values suppress the use of the information present in the topology file. Positive values add to the information that was in the topology file. If NBXMOD is not specified for a command, it defaults to 2.
The third way to change exclusions is the use of the EDIT command, see section EDIT -- Edits the PSF and Hydrogen Bonds).
DISUlfide [NBXMod int] NCYST (I5) IRES,JRES (2I5) repeated NCYST times
This command requires formatted (not free field) input following the DISULFIDE line.
The first line following the command gives the number of cystine cross bridges to be made; The lines following give the residue numbers (not residue identifiers) of the cysteines to be linked. The residue number of a residue is its position in the list of all residues in the structure including any special termini which have been added. The linkage process involves adding bonds, bond angles, torsion angles, and non-bonded exclusions for the additional bond. The NBXMOD option controls the automatic generation of non-bonded exclusions, see section NBXMOD -- Automatic Generation of Non-bonded Exclusions, for more details.
An attempt has been made to ease the burden of going from residue identifiers to residue numbers which will be different for segments which have N-terminal residues added. Whenever the type of segment as specified in the READ SEQUENCE command is CHARMM explicit hydrogen or all hydrogen, the residue numbers will first be increased by one. If the two residues given are not cysteines, the residue numbers will be decremented by one, and the attempt repeated. If this fails, the command will die. Note that this does not help you if you have more than one segment in your structure.
When using disulphides, it is important that the sequences reference a cysteine residue which is intended to be joined. In an all atom topology file, see section AMBER94RTF, there are two cysteine residues, one of which has a thiol and one of which has a sulfur.
It is not possible to bridge an atom in the primary space with that of a symmetric image using this command (see section Symmetry and Molecular Images).
PATCh HEME [NBXMod int]
histidine-heme-spec
or
PATCh LIGA [NBXMod] int]
carbonmonoxide-heme-spec
PATCH HEME is used to patch the ligation of a histidine to a
heme residue. The histidine-heme-spec is a pair of integers read by a format
of 2I5,
IRES and JRES, are the residue numbers of the histidine and heme,
respectively. The bond is formed between the NE2 of the histidine and FE
of the HEME. For each bond formed, additional bond angles, torsion
angles, and non-bonded exclusions are added.
PATCH LIGAND is used to patch the ligation of a carbon
monoxide to a heme. The carbon monoxide-heme-histidine-spec consists of
three integers, read by a format of 3I5, and again are not free
format. The three numbers refer to the histidine, heme, and CO residue
numbers respectively. This patch takes care of the carbon monoxide heme
bond. It should be called after PATCH HEME is called.
The NBXMOD option controls the automatic generation of non-bonded exclusions, see section NBXMOD -- Automatic Generation of Non-bonded Exclusions, for more details.
WARNING: This code has been tested only with the extended atom topology file. It may not work for current editions of the other amino acid topology files, and it will not work if there is a proton placed on the NE2 of the histidine.
It is not possible to patch any interaction involving atoms of a symmetric image using these command (see section Symmetry and Molecular Images).
EDIT [NBXMod int] edit-commands
edit-commands are described below. They must be terminated with an END command. The edit-commands are not free field.
EDIT is used to edit the PSF and also the a hydrogen bond list without explicit hydrogens. The following operations are possible: Any bond, bond angle, torsion angle, improper torsion angle, hydrogen bond donor, hydrogen bond acceptor, non-bonded exclusion, or hydrogen bond may be deleted or added. In addition, the parameter type code, charge, and IUPAC name of any atom may be changed.
The operations with hydrogen bonds and with hydrogen bond donors and acceptors are obsolescent as one cannot add the proton to any new hydrogen bonds, nor can one add the various antecedents to the hydrogen bond donor. At some point, this will be fixed.
The edit-commands are all fixed format commands. Each command
except the END command consist of three parts. First, one specifies an
alphabetic command using words read with a 2(A4,6X) format. The
following line consists of single integer using (I5) format giving the
number of changes. Finally, that number of lines follows, where each
line specifies one change.
The END command consists of the word END in the first of column of a line with nothing else on the line. This terminates the EDIT command.
The NBXMOD option controls the automatic generation of non-bonded exclusions, see section NBXMOD -- Automatic Generation of Non-bonded Exclusions, for more details.
{ ADD } { BOND } (2A10) Fixed format
{ DELEte } { THETa }
{ PHI }
{ IMPHi }
{ HBONd }
{ DONOr }
{ ACCEptor }
{ NONBond }
NCHANG (I5)
NRESI,ATOMI,NRESJ,ATOMJ,NRESK,ATOMK,NRESL,ATOML (4(I5,1X,A4))
repeated NCHANG times
Most of the keywords are self explanatory. In the case on NONBOND,
the nonbonded exclusions list is changed, not the actual list of
nonbonded interactions. NCHANG is the number of elements to be added or
deleted. NRESI specifies the residue number of the first atom, and ATOMI
specifies its name. For example, deleting a peptide bond between the
fourth and fifth residue would be specified as
4 C 5 N
If a number of different internal coordinates are to be changed, separate ADD or DELE commands must be used. They can appear in any order. One must specify only as many atoms as there are in the interaction. For hydrogen bonds, only two atoms may be specified even though the proton may be explicit represented.
WARNING: This routine does not make the correct checks if you make a mistake. If you specify an atom incorrectly, or if the interaction you wish to delete is not there, the results are unpredictable. At some point, this may be fixed.
MODIfy (A4) Fixed format NCHANG (I5) NREST,ATOMT,ICODE,ANAME,CHARG (I5,1X,A4,I5,1X,A4,F10.3) repeated NCHANG times
NCHANG is the number of atoms to be modified. A card with the changed
values must be read in for each atom to be changed (in any order). NREST
and ATOMT are the residue number and atom type of the atom to be
changed. ICODE, ANAME and CHARG are the new chemical type code
(IAC array entry), new atom type and new atomic charge given to
the atom. If a field is left blank, the old value is retained. Editing
of this type could be used, for example, to change the carbonyl oxygen
of a terminal residue to an atom with attributes corresponding to a
carboxyl oxygen (in the standard execution of the program, modifications
like this are done automatically if the proper type is specified with
the sequence).
Go to the previous, next section.