Dear Vidya,
On Tue, Apr 07, 2009 at 02:20:42PM +0100, Vidya Chandran wrote:
> Dear All,
>
> I am trying to use getax for finding NCS in a map radius of 100A and get
> the error 'ERROR: more than MAXPTS points in sphere - please recompile'.
I guess you're using a sphere of radius 100A for searching for the
NCS?
The error you're getting isn't directly related to the map/grid size,
but to the generated simple description for the region following your
NCS. Some information:
1. you usually don't need to use a map calculated at high resolution -
most of the time something at 6A is good enough. This already
reduces the grid size of your map and has a direct effect on the
number of points used for that simple description
2. what are you trying to search for? Have you got some idea what the
multimer should look like? Is it a 2-fold, 3-fold etc (i.e. Cn
symmetry) or a more complex multimer (Dn symmetry)?
I would first start looking for the low-order symmetry axes
(2-fold, 3-fold).
3. what shape do you expect the multimer to be?
How large do you expect the volume to be that follows the n-fold
symmetry? With a radius (not diameter) of 100A this would be a VERY
large dimer.
But maybe you have something really big that that looks more like a
cage (like in the FAS case)? Then you would have a 'shell' of
protein with a large cavity in the middle: in which case you could
use something like
SPHERE 100.0 80.0
to create such a shell (outer and inner radius given). This will
also reduce the number of points needed to describe the multimer.
4. instead of a large shell/sphere: just use a small SLICE with the
desired radius but a smaller height (e.g. 20A high). This will also
reduce the number of points needed.
In any case: if you run searches for Cn symmetry (i.e. proper n-folds)
make sure to visually inspect the resulting correlation maps. You'd
expect long stretches of 'density' showing you the location of the
n-fold rotation axis.
> I am running CCP4 6.0.2 on my intel macbook pro. I am not sure which
> parameter to change in the getax.f file and how to exactly recompile
> getax with the existing CCP4 installation. Any advice would be really
> appreciated.
If you really want/need to recompile: change the line
parameter (maxpts=20000)
in $CCP4/src/getax.f and then
% cd $CCP4/src
% make getax
Note: this will only work if you did the compilation of CCP4 yourself
_and_ you left all files in place afterwards.
Cheers
Clemens
>
> Thanks,
> Vidya
--
***************************************************************
* Clemens Vonrhein, Ph.D. vonrhein AT GlobalPhasing DOT com
*
* Global Phasing Ltd.
* Sheraton House, Castle Park
* Cambridge CB3 0AX, UK
*--------------------------------------------------------------
* BUSTER Development Group (http://www.globalphasing.com)
***************************************************************
