STEREO (CCP4: Supported Program)
- Extract coordinates from stereo diagrams
program recreates three-dimensional coordinates on the basis of a pair
of two-dimensional ones extracted from a stereo diagram. The input
coordinates are taken from the logical name XYZIN (see
FORMAT OF XYZIN)
into a PDB-format file on XYZOUT. The output will not necessarily be
a sensible PDB file, especially if the optional information about atom
names and residue types isn't supplied. The programs
may be useful for manipulating the output.
The input coordinates are expected to be measured (from some
publication for which you don't have the coordinates electronically)
with a flatbed cross-hair digitiser or similar device. If you don't
have such a device you might be able to do this:
make a transparency of the picture;
stick it against a graphics screen and fire up some drawing program
from which you can get the data in a sensible form, e.g. `xfig';
draw points or lines to trace each stereo view and save the drawing;
mangle the drawn points output into a suitable form for input to this
program (see below) using your favourite utility.
Available keywords are:
Angular separation of the stereo components in degrees. If not given
the correct value will be searched for.
Viewing distance in inches. If not given the correct value will be
Criterion for removing bad measurements from <phi> and <v> searches.
The default value is 1.5. Prior to <phi>, <v> search, the program
uses <phi>=3 degrees, <v>=infinity, and distances used in constraints.
All distances which are outside the range (s*<crit>,s/(<crit>**2)),
where s is the standard Ca-Ca distance of 3.84 A, will be rejected
from the search.
CONSTRAINT <const1> <const2> <const3> <const4>
Specify a constraint. If no constraints are specified a polypeptide
chain of alpha carbon atoms will be assumed and the distances between
succeeding pairs of atoms will be refined against the standard alpha
carbon - alpha carbon distance of 3.84 Angstroms.
First i-th atom in constraint
First j-th atom in constraint (if negative, will count backwards).
Last i-th atom in constraint
Distance s(i,j) between i-th and j-th atoms measured in Angstroems.
If <const2>=0 for the k-th constraint line, an alpha helix is assumed.
In this case <const1> should be the first carbon alpha atom in the
helix (call it atom i) and <const3> should be the last carbon alpha
atom in the helix (call it atom j) and a series of constraints is
generated in place of the given constraint as follows.
- The first set of constraints will be for
atom i to atom i+2
atom i+1 to atom i+3
etc. until i+n=j
- The second set of constraints will be for
atom i to atom i+3
atom i+1 to atom i+4
etc. until i+n=j
the last constraint will be for
atom i to atom j
By assuming the parameters for an alpha helix, the program works out
the appropriate values for s(i,j) in subroutine HELIX.
Examples of particular constraints that may be set are
An alpha carbon chain of n atoms.
CONSTRAINT 1 2 n-1 3.84
Two antiparallel strands, each n residues long, starting at residue
numbers p and q.
CONSTRAINT p -(q+n-1) (p+n-1) 4.72
(Note: there are two different Ca-Ca distances between antiparallel
strands and this number is an average.)
An alpha helix n residues long starting at residue p.
CONSTRAINT p p+4 p+n-4 6.20
The program will minimise the sum over all the constraints of
FORMAT OF XYZIN
The input data file comprises free format records of the form
<xl> <yl> <xr> <yr> [ <atom name> [ <residue name> [ <residue number>] ] ]
These are the left and right stereo diagram coordinates. It is
assumed that the coordinates will be given in microns, so as to
provide the viewing distance in inches. The program will work if the
coordinates are in any other units but the <v> distance will not be in
inches but on a corresponding arbitrary scale. Each coordinate has an
optional <atom name>, <residue name> and <residue number> which will
be used in he XYZOUT file if given. Otherwise dummy tags will be
The output is hopefully self explanatory but essentially it consists
of the following sequential items following header.
M. G. Rossmann March 1979
Dept. of Biological Sciences
West Lafayette, Indiana 47907
Modified and reorganized with new math routines
email@example.com August 5, 1994
Hacked into CCP4 conventions (basically i/o) at Daresbury.