[gmx-users] Rigid body optimization

[David]

On Thu, 2003-12-04 at 22:06, Mengjuei Hsieh wrote:
> Hi,
> 
> I want to confirm that the all-angles constraints in gromacs is almost not
> usable. I tried lots of times trying to figure out. The problem might be
> related to whether or not the starting structure is built with ideal
> geometries. This might not be a bug, but a feature instead. Angle
> constraints are very complicated and very hard to iterate to reach under the
> tolerance if we use SHAKE/LINCS algorithm.
> 
> Perhaps we need constraints according to the starting structure, not ideal
> angles/lengths like what CHARMM does.
If you want to do it nicely, you'd have to sum the atomic forces over
the molecule and then integrate the whole molecule as a rigid body
indeed (with net force and torque giving exactly six degrees of
freedom). It wouldn't be so hard to implement, except when molecules are
split over processors. I assume as well it would require double
precision. For the sequential bit, I would gather that it would be less
than 100 lines of code to add in update.c


> 
> Any idea?
> 
> Best,
-- 
David.
________________________________________________________________________
David van der Spoel, PhD, Assist. Prof., Molecular Biophysics group,
Dept. of Cell and Molecular Biology, Uppsala University.
Husargatan 3, Box 596,  	75124 Uppsala, Sweden
phone:	46 18 471 4205		fax: 46 18 511 755
spoel at xray.bmc.uu.se	spoel at gromacs.org   http://xray.bmc.uu.se/~spoel
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++