Umbrella sampling using colvarconfig file with dihedrals

From: Aaron Larsen (
Date: Wed Feb 05 2014 - 11:07:23 CST


I am attempting to perform umbrella sampling with a harmonic restraint on a
dihedral angle of a nucleoside. Minimization sand unrestrained simulations
are working well but I am having some difficulty proceeding onto umbrella
sampling. I have read through the manual but there is not a lot of
information regarding this process using dihedral angles in particular.

The dihedral angle that I am attempting to restrain is O4'-C1'-N1-C2,
otherwise known as the chi angle. This is the angle which describes the
relative twist between the nuceobase and the sugar. I have prepared a
modified version of my NAMD config file use for the simulation to include a
term for colvars.

The corresponding terms that I have placed in my config file are:

colvars on

In the colvars config file that I have pointed to, I have attempted to
define the colvar corresponding to my dihedral angle with the atom numbers
that correspond to the correct atoms in the appropriate pdb file. This is
the contents of the colvarconfig file are:

#Colversconfig file for diheral angles
colvarsTrajFrequency 100 # output
values every 100 steps
colvarsTrajAppend off
analysis off
colvar {
name chicolvar # chose a name to identify the variable
outputSystemForce yes
dihedral {
group1 {
atomNumbers 8
group2 {
atomNumbers 7
group3 {
atomNumbers 15
group4 {
atomNumbers 13
atomsFile ../common/U_wb.pdb
atomsCol O
atomsColValue 2.0
harmonic {
name chiharmonic #the name of your potential
colvars chicolvar
centers 0.0
forceConstant 10.0

My questions are as follows:

1) Is it correct that the harmonic block be in the colversconfig file
and not in the NAMD config file? Is there any additional information that
should be in the NAMD config file other than simply the colvars block?

2) Is it correct that the value of 'centers' should be a single
decimal corresponding to the desired dihedral angle that I would like
center the constraint on?

3) Is it correct to assume that the force constant will be in units
appropriate to a dihedral angle? In this case kcal·mol-1·deg-2?

4) When attempting to run the simulation, I am getting an error
indicating that the term 'atomsFile' is not recognized by NAMD. Is this
term no longer in use?

 Thank you very much for any input that you have.


Aaron Larsen, Ph.D.
Harvard University Department of Chemistry and Chemical Biology
Harvard Medical School Department of Genetics
Mobile: 617-319-3782
FAX: 617-643-3328

This archive was generated by hypermail 2.1.6 : Wed Dec 31 2014 - 23:22:04 CST