RE: improper angle

From: JC Gumbart (
Date: Sat Nov 12 2011 - 01:14:09 CST

This is also covered in the appendices of the NAMD tutorial.


From: [] On Behalf Of johan strumpfer
Sent: Friday, November 11, 2011 6:22 AM
To: Marco Sant
Subject: Re: namd-l: improper angle


Dear Marco and Bin,

Looking at the code - the angles is between the planes 1-2-3 and 2-3-4. I presume also that when you ask about the restriction, you're referring to CHARMM-formatted parameters files for use in namd. Indeed there is a restriction, the order of the atoms must match the order they are found in the IMPROPER entry of the psf file. If I remember correctly, the convention is that the central atom is in position A in A-B-C-D.


Johan Strumpfer: <>
Theoretical and Computational Biophysics Group
3115 Beckman Institute
University of Illinois at Urbana-Champaign
405 N. Mathews
Urbana, IL 61801, USA

On Fri, Nov 11, 2011 at 3:34 AM, Marco Sant <> wrote:

Dear all,
if in the parameters file I specify an improper angle
as formed by the following atoms:


does it mean that the improper angle is the one
between the planes A-B-C and B-C-D ??

I guess so looking at the meaning of dihedral angle,
A-B-C-D, as angle between planes A-B-C and B-C-D,
and seeing in the source code that dihedrals and impropers
have a similar approach.

Still, I have to answer to a referee for the publication
of an article and I need a RELIABLE answer.
I apologize if this is explained somewhere and it is
just me that cannot find it!

Secondary aspect. In an improper angle there
is always a central atom to which other three atoms
are connected.
In the lower schema the central atom is numbered as 1.

     / \
   2 4

Is there a restriction, when I specify the potential for the
improper A-B-C-D, to which position the central atom
should have?
For example, in my calculations I specify the improper
of the above schema as: 2-3-1-4.
In other words: should the central atom 1,
appear in position B and C, or in position A and D,
when specifying A-B-C-D improper?

Thanks a lot,
Marco and Bin


Wireshark Cookie Dump:
[ ]

Wireshark Cookie Dump:
[ ]

This archive was generated by hypermail 2.1.6 : Mon Dec 31 2012 - 23:20:59 CST