| GRID manual | ||
|---|---|---|
| <<< Previous | Next >>> | |
Chapter 6. User guide to programme GRIN
Programme GRIN will now be described in detail. New Users are advised to skip this detailed description until they need to study it. They should start by looking at the tutorial examples.
6.1. Structure of programme GRIN
The relationships between Files and Programmes are shown in Figure 2.
Programme GRIN requires two input files (PDB: Figure 1, and GRUB: Figure 5) and may also need a few input directives. One directive, IEND, must always be supplied.
GRIN produces two output files: GRINLOUT (Figure 4) is the lineprinter output with warnings and messages which should be carefully considered. GRINKOUT (Figure 6) is an ASCII file for input to Programme GRID. In certain circumstances GRINKOUT must be edited, but this is not normally recommended.
6.2. Input and output channels
Five "Logical Units" are used by Programme GRIN. These are the Channels through which data, results, and other information enter and leave the Programme as shown in Figure 2. Two are used for the input files PDB and GRUB, and two more are for the output files GRINLOUT and GRINKOUT. The fifth channel is for the command file grin.in with the Directives for GRIN.
The five channels are:
| INKO | is the input channel for the coordinates of the atoms in the Target molecule which is to be studied. The file (PDB) which contains these coordinates must be in Protein Data Bank (Brookhaven) format. Alternatively, when several different Targets are to be processed one after the other as a Set, a list of file names will be accepted by Channel INKO instead of one single PDB file name. Programme GRIN as supplied on CD has a normal default value INKO=11 |
| INAT | is the input channel for file GRUB which contains the Energy Variables appropriate to each type of atom in the chosen Target molecule. Normal default value INAT=10 |
| LOUT | is the output channel for a lineprinter GRINLOUT. All the input directives are copied to the lineprinter for checking purposes, and they are followed by a list of warnings and messages concerning the input file PDB and datafile GRUB. Normal default value LOUT=6 |
| KOUT | is the output channel for the merged file GRINKOUT which is to be used as input to the next Programme GRID. Normal default value KOUT=20 |
| IDIR | is the input channel for the directives themselves. It may therefore be used to change the above channels and all the other directives at run time. This value is normally IDIR=5 when the Programme is supplied |
The correct use of the input and output channels with various operating systems is described in more detail below.
6.3. Input to programme GRIN
Input consists of three components: the datafile GRUB (Figure 5); the directives; and the PDB file (Figure 1). Alternatively a list of file names may be input instead of the one PDB file.
6.3.1. Input conventions
Before considering the input to Programme GRIN in detail, it is necessary to describe certain conventions which have been defined by the Protein Data Bank or elsewhere. The following conventions are in common use and will be followed in this User Manual.
6.3.1.1. Atoms and hetero-atoms
The conventions adopted by the Protein Data Bank allow it to contain information about small molecules such as drugs, and macromolecules such as proteins or nucleic acids. It is a characteristic of these macromolecules that they are composed of many small structural units called residues. However the number of different kinds of residue is strictly limited. Thus proteins contain some twenty different kinds of amino-acid residues, and these have been given a special status in the conventions of the Protein Data Bank. The atoms of a residue are designated ATOMS, whereas the atoms of most other molecules are designated HETERO-ATOMS (which is abbreviated to HETATMS).
An ATOM record in a Protein Data Bank File is quite distinct because it begins with the six characters 'ATOM ' like this:
ATOM 1 CA LYS 1 1.933 20.601 21.051 |
HETATM 1 C1= MOL 1 -1.750 -2.250 0.000 |
6.3.1.2. Atoms and extended atoms
The X-ray observations of a macro-molecular structure may not define the positions of all the atoms, and it frequently happens that the coordinates of the hydrogen atoms are not recorded in the Protein Data Bank. It is then convenient to treat some chemical groups (e.g the methyl group CH3) as if it were a single "Extended Atom", instead of a cluster of one carbon and three hydrogens. This is particularly advantageous for energy calculations, because it is only necessary to consider the one extended atom with three coordinates, instead of the four individual atoms with twelve.
In Programmes GRIN and GRID an Extended Atom consists of one heavy atom and any attached hydrogens. The organisation of any lone pair electrons is also considered. Thus the aliphatic hydroxyl group of the amino-acid serine is treated as an Extended Atom having one sp3 oxygen with two lone pairs and one bonded hydrogen. Most of the ATOM entries in datafile GRUB relate to residues (e.g. amino-acids or monosaccharides) which combine to form macromolecules. The hydrogen positions in these macromolecules are not normally observed, and so these ATOM entries are parameterised as Extended Atoms. However, information is also given in Datafile GRUB about the hydrogens which are subsumed in each Extended ATOM, and this information is used if hydrogen coordinates are actually supplied.
There is a separate list of HET variables at the end of Datafile GRUB for use with small molecules. In contrast to the ATOM variables for macromolecules, almost all of these HET variables are parameterised as individual HETATMS. In only a very few cases do the HET variables refer to Extended Atoms, and those particular HETATM entries are specially marked with the letters EA in Datafile GRUB.
It should be noted that some workers use "United atom" as an alternative name to "Extended atom".
| <<< Previous | Home | Next >>> |
| Introduction to programme GRIN | Up | Programme modes |