[gmx-users] constraint/restrain distance

[Ángel Piñeiro Guillén]

Thanks for your answer Berk,

I have tried quite different distances (even distances at which the
molecules are too far to interact) and the simulations still crash. It seems
that the probability of crashing does not depend on the distance. Moreover
the average force for the longer distances before crashing is cero. The
constraint algorithm seems to be working well (for a while which in some
cases takes several nanoseconds) and the forces are reasonable for each
distance... or at least the dependence of the forces on the distance is
reasonable. I understand that to constraint the distance between the center
of mass of two molecules lincs has to "move" one of the molecules a certain
distance after each timestep keeping its structure (probably it has to do
the same with some water molecules). Is there any particular case in which
lincs can fail?

In regard to the AFM and umbrella pulling I understand that they have been
implemented with different purpose but by using them with the right
parameters they should provide the same force (I guess)

Angel Pineiro.



----- Original Message -----
From: "Berk Hess" <gmx3 at hotmail.com>
To: <gmx-users at gromacs.org>
Sent: Thursday, August 26, 2004 5:10 PM
Subject: RE: [gmx-users] constraint/restrain distance


>
>
>
> >From: ÿffffc1ngel Piÿfffff1eiro <anxomx at yahoo.es>
> >Reply-To: Discussion list for GROMACS users <gmx-users at gromacs.org>
> >To: gmx-users at gromacs.org
> >Subject: [gmx-users] constraint/restrain distance
> >Date: Fri, 20 Aug 2004 14:39:52 +0200 (CEST)
> >
> >Dear gromacs users,
> >I want to calculate the force to fix the distance
> >between two molecules A and B. I think that it is not
> >possible at the moment to restraint the distance
> >between different molecules in the topology, so I
> >tried to use the pull code. I started to play with the
> >3 different options (AFM, CF and umbrella sampling)
> >but I found several problems.
>
> The crashes in constraint pulling could be due to the short distance.
> As I understood the two small molecules are at short distance.
> The constraint could introduce a high stiffness into your system
> such that with some conformations of the two molecules the forces
> get too high.
>
> As for AFM and umbrella pulling.
> It would be nice to use these for this purpose.
> But they have been implemented with a different purpose.
> The problem is that the distance between the group and the reference
> is set with a vector, not with a distance.
> So you can restrain the vector between the centers of mass of the two
> molecules
> to (0.3,0,0) by setting:
> pulldim = Y Y Y
> pos1/afm_init1 = 0.3 0 0
> Using pulldim = Y N N leaves the molecules free to move in the y and z
dir.
>
> It might be better to change the afm and umbrella options to the way
> the constaint option works, with a direction that can be zero
> so you can restrain the distance and not to a vector.
>
> Berk.
>
> _________________________________________________________________
> Protect your PC - get McAfee.com VirusScan Online
> http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963
>
> _______________________________________________
> gmx-users mailing list
> gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-request at gromacs.org.