Previous Thread
Next Thread
Print Thread
Page 2 of 2 1 2
Joined: Mar 2019
Posts: 14
A
Forum Member
OP Offline
Forum Member
A
Joined: Mar 2019
Posts: 14
I have been trying to use shake to constrain the bond distances of my target molecule, however this does not prevent the bond angles from changing. When I try shake angle then all the atoms in the molecule move wildly.

It is baffling me that I am unable to find a function in CHARMM to perform a task that is not uncommon (preventing any changes to a molecule's shape).

Am I looking at the wrong commands? I have tried:

cons dihe
cons harm abso
cons IC

shake bond
shake angl

Joined: Sep 2003
Posts: 4,840
Likes: 3
Forum Member
Offline
Forum Member
Joined: Sep 2003
Posts: 4,840
Likes: 3
Which biomolecular MD programs have a function to prevent any changes to a molecule's shape, and how do these function differ from the constraints and restraints available in CHARMM?

You have to remember that the main purpose of MD is to change the conformation...

All atoms in the molecule should not move wildly when you SHAKE ANGLES if the angles in the structure are close to what SHAKE tries to impose; I have however never used SHAKE ANGLES myself.


Lennart Nilsson
Karolinska Institutet
Stockholm, Sweden
Joined: Sep 2003
Posts: 8,597
Likes: 12
rmv Offline
Forum Member
Offline
Forum Member
Joined: Sep 2003
Posts: 8,597
Likes: 12
The task is common, but not to the tolerances you seem to require. The concerns are more typically preserving dihedral states and chirality with restraints during equilibration, and relying on the balls and springs force field for the rest.

One option might be defining custom atoms for your target molecule, so that you can increase the force constants for the bond, angle, and dihedral terms to achieve the degree of stiffness you want. However, this would require a non-trivial effort.

As far as I know, only the SHAPES facility was intended for rigid body dynamics, but I don't know that much about it.

If you wish to allow whole molecule translation and rotation, CONS HARM BESTFIT might be an option instead of CONS HARM ABSO, but I doubt it would achieve the tolerances you want.


Rick Venable
computational chemist

Joined: Mar 2019
Posts: 14
A
Forum Member
OP Offline
Forum Member
A
Joined: Mar 2019
Posts: 14
GROMACS has this function, however that software is designed for proteins. TurboMole has this function, but is not a biomolecular MD suite.

Lennart, that wasn't a very helpful remark. I am aware that the purpose of MD is to change the conformation... there are also times when you want to change only a portion of the conformation and keep another portion rigid.

rmv: I will try SHAPEs. I checked out that link you provided and am still figuring out the syntax for my needs. I have already tried CONS HARM BESTFIT and didn't get the desired effect.

Joined: Sep 2003
Posts: 8,597
Likes: 12
rmv Offline
Forum Member
Offline
Forum Member
Joined: Sep 2003
Posts: 8,597
Likes: 12
CHARMM was also designed for biopolymers, (years before GROMACS), but rigid body dynamics has not been pursued very much.


Rick Venable
computational chemist

Joined: Mar 2019
Posts: 14
A
Forum Member
OP Offline
Forum Member
A
Joined: Mar 2019
Posts: 14
Instead of trying to fix the molecule in place during step6_equilibration, could I let those steps run unconstrained, do one step in production and then edit the coordinates/velocities from my gas phase dynamics into the restart file? I realize that this will require some efforts with positioning in the membrane.

My other idea is read the initial coord/velocities into comparison, do the equilibration and then read the comp set back into the system. Would this work?

Page 2 of 2 1 2

Moderated by  BRBrooks, lennart, rmv 

Link Copied to Clipboard
Powered by UBB.threads™ PHP Forum Software 7.7.5
(Release build 20201027)
Responsive Width:

PHP: 7.3.31-1~deb10u1 Page Time: 0.016s Queries: 27 (0.011s) Memory: 0.7617 MB (Peak: 0.8363 MB) Data Comp: Off Server Time: 2022-07-03 12:11:42 UTC
Valid HTML 5 and Valid CSS