Re: using CHARMM-format Amber parameters in NAMD

From: Basheer (
Date: Wed Aug 20 2014 - 22:15:22 CDT

Thank you so much for all your input! Things are much clearer now (yes, Kenno is right; I was confused about topology data within psf files).

I will now attempt to generate the psf (of some short dsDNA segments) using the correct parameter inputs, and I will let you know if I have any questions.

I'm very grateful for everyone's input. :)

- Basheer Subei

On Aug 20, 2014, at 10:27, Kenno Vanommeslaeghe <> wrote:

> Just wanted to add that the topology information is fed to NAMD through the psf file; many beginning users don't realize this. So the psf file needs to be generated using the correct topology (rtf) file, i.e. the one that comes with the parameter (prm) file you're feeding NAMD. It appears to me that Basheer is trying to combine a psf file for the CHARMM FF with and AMBER FF parameter file, which will never work.
> On 08/20/2014 12:34 AM, Mike Makowski wrote:
>> Hi Basheer,
>> You're certainly right in assuming that the AMBER to CHARMM format files
>> are compatible with CHARMM type coordinate files. In fact, NAMD allows
>> input of AMBER parmtop files with the use of pdb as well. You problem
>> results from incompatibility between your topology and parameter files.
>> The parameter file has to define all necessary
>> bonding/nonbonding/electrostatic/etc. parameters for all atom types
>> specified in your topology file. If CN7 isn't defined in the AMBER
>> parameter file, it's likely that it uses a different atom type to
>> represent that particular atom. More specifically, browse through your
>> topology file for CN7. Here you define that atom but in the AMBER
>> parameter file, if you perform a search, you'll notice it's not to be
>> found. One of the troubles of using these AMBER or OPLS files is
>> constructing topologies that are compatible with the parameter file. Hope
>> this helps. Good luck!
>> Mike
>> Michael Makowski
>> University of California, Irvine
>> Department of Chemistry,
>> Chemical and Material Physics,
>> Irvine, CA 92617
>> Cell: 267 588 9865
>> Office: 949 824 5776
>> On Tue, Aug 19, 2014 at 5:16 PM, Basheer Subei <
>> <>> wrote:
>> Hello all,
>> *How do I use the Amber force field parameters in NAMD in the CHARMM
>> format?* I found some CHARMM-format Amber parameter and topology files
>> in the "non_charmm" directory in the c32b1 version of CHARMM ff. I
>> also found some online here
>> <>,
>> listed under "AMBER99 for use in CHARMM", and I tried using those
>> parameter files in my NAMD configuration file (.namd), but it would
>> complain about not finding "VdW parameter for atom type CN7".
>> _*Specifically, what I did is:*_
>> - I followed the NAMD nanopore tutorial
>> <>
>> up to section 3.1.9 (for minimizing ssDNA in NAMD).
>> - In min.namd, I replaced the CHARMM "par_all27_na.prm" file with the
>> "par_amber_cornell.inp" (the Amber-in-CHARMM-format parameter file).
>> ATOM TYPE CN7". It works fine when I don't replace the parameter file
>> (so I've ruled out any problems with not following the tutorial
>> correctly).
>> I tried using a few variants of these parameter files I found online
>> (Amber force fields but CHARMM format), but they all give the same
>> error. Now that I'm inspecting the actual files and comparing them to
>> the default CHARMM parameter files in c32b1, I notice they're
>> significantly different. And mailing list threads
>> <>
>> about this topic are confusing me.
>> What am I missing? Can someone please point me in the right direction?
>> Am I right in assuming it is possible to use these
>> Amber-in-CHARMM-format parameter files with structures generated using
>> CHARMM (psf and pdb) in NAMD?
>> Sorry if some of this makes no sense. Thank you for your time!
>> - Basheer Subei

This archive was generated by hypermail 2.1.6 : Wed Dec 31 2014 - 23:22:44 CST