Re: ligand flies away from binding pocket

From: Mert Gür (
Date: Fri Feb 13 2009 - 13:34:58 CST

Dear Osman,
This can be done by applying an constraining force. Altough this bias must
somehow be subtracted when the energy is calculated.
Lets first generate and reference file for the constrained atoms by setting
their occupancy column to the spring constant of the force.

set all [atomselect top all]
$all set occupancy 0

set A [atomselect top "protein and backbone"]
$A set occupancy 20
$all writepdb refres_20.pdb

and then adding these to your configuration file
      constraints on
 consexp 2
 consref refres_20.pdb
 conskfile refres_20.pdb
 conskcol O
 selectConstraints on
 selectConstrX on
 selectConstrY on
 selectConstrZ on
I hope this helps.

On Thu, Feb 12, 2009 at 7:39 PM, Osman Yogurtcu <>wrote:

> Hi,
> I have a 300 K water NPT simulation of a ligand (one aminoacid) bound to a
> receptor. After I do minimization and equilibration (for 1 ns), I do an MD
> run. I observe that my ligand leaves the binding pocket after 2 ns.
> My question: is there a trick that I can make sure that my ligand sits in
> the binding pocket all the time or should I define bonds artificially and
> fix the ligand?
> thank you,
> Osman

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:52:22 CST