Re: Constant Force Pull

From: McGuire, Kelly (mcg05004_at_byui.edu)
Date: Thu Dec 13 2018 - 10:33:13 CST

Next message: Francesco Pietra: "Running QM-MM MOPAC on a cluster"
Previous message: L-: "run MD simulation in NAMD using stages file strategy or all in one file ?"
In reply to: JÃ©rÃ´me HÃ©nin: "Re: Constant Force Pull"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]

Yep, sorry about, clicked the wrong reply. It sounds, though, that you wouldn't necessarily use any restraints on the system, and that seeing the membrane pull apart and the water box expanding when the drug is has a force on it, is a sign that the setup is wrong, correct?

Kelly L. McGuire

PhD Candidate

Biophysics

Department of Physiology and Developmental Biology

Brigham Young University

LSB 3050

Provo, UT 84602

________________________________
From: Jérôme Hénin <jerome.henin_at_ibpc.fr>
Sent: Thursday, December 13, 2018 8:39:28 AM
To: McGuire, Kelly
Cc: Namd Mailing List
Subject: Re: namd-l: Constant Force Pull

Please keep the conversation on-list.

There are several possible ways to go about this. One way is to fix some atoms, I suppose, but better use NAMD "constraints", which are actually restraints. A misnomer that I hope will be corrected eventually.

Another way is to use the Colvars Module to apply a force on an internal coordinate of the protein-ligand system.

Jerome

On Thu, 13 Dec 2018 at 16:18, McGuire, Kelly <mcg05004_at_byui.edu<mailto:mcg05004_at_byui.edu>> wrote:
Interesting, so it is not required to fix any atoms? The force, if set up correctly, should only pull on the ligand and the rest of the system should behave normally? I use constantforce yes, and consforcefile myfile.ref, and within myfile.ref only the atoms for the ligand have a value in the occupancy column, the others are zeroes, and the x y z of the ligand atoms are set to 0.00 0.00 1.00, so that they are pulled in the positive z direction, the other atoms' coordinates were left unchanged.

Kelly L. McGuire
PhD Candidate
Biophysics
Department of Physiology and Developmental Biology
Brigham Young University
LSB 3050
Provo, UT 84602

________________________________
From: Jérôme Hénin <jerome.henin_at_ibpc.fr<mailto:jerome.henin_at_ibpc.fr>>
Sent: Thursday, December 13, 2018 6:06:12 AM
To: Namd Mailing List; McGuire, Kelly
Subject: Re: namd-l: Constant Force Pull

My advice would be to just pull on the ligand with a counter-force on some protein atoms, distributed around the binding site. There should be no need for any restraint on the membrane, that's a symptom that something else is very wrong.

Jerome

On Thu, 13 Dec 2018 at 10:48, McGuire, Kelly <mcg05004_at_byui.edu<mailto:mcg05004_at_byui.edu>> wrote:

I am trying to do a constant force pull on my ligand, pulling it out of a protein. I have the backbone atoms and certain atoms in the DMPC membrane constrained. The force on the ligand is 1 kcal/mol/A^2 or 69.479 pN A. As the ligand is moving, the waters on both sides of the membrane expand away from the membrane the simulation stops with the error:

ERROR: Constraint failure in RATTLE algorithm for atom 60956!
ERROR: Constraint failure; simulation has become unstable.

Before I put a constraint on the membrane, it was separating in opposite directions along the z-coordinate. As for the water, I can't constrain those or the simulation stops with periodic boundary errors. How do I do a constant force pull on a ligand in a protein/membrane/water system?