[gmx-users] charmm c36 lipids

[Pär Bjelkmar]

Hello Sven, Drew and gmx-users,

I've gotten requests from users that want to use the c36 CHARMM lipids in GROMACS (see below). So I answer both of you and the rest of the community in this e-mail. 

I have a script that parses the top and par files of CHARMM force field(s). It's not the most general piece of code there is so it will not be able to convert files that do not follow the format the CHARMM top and par files had when I wrote it (version c32b1). Also, it cannot parse the toppar files (i.e. what's called the toppar stream files in CHARMM). I downloaded the force field from:
http://mackerell.umaryland.edu/download.php?filename=CHARMM_ff_params_files/toppar_lipid_all36.tgz

My script is able to parse the top_lipid_all36.rtf and par_all36_lipid.prm files without any errors and generate the corresponding GROMACS rtp, atp, bonded and nonbonded files. I'm not an CHARMM expert but if they haven't changed the force field considerable lately (by changing parameter definitions and functional forms of the force field for example) the files generated by the script should be the correct GROMACS translation. However, I cannot guarantee that all went well so I would suggest that those of you who are interested in this look through the files and convince yourself that the script did the job before running production runs with these parameters. 

I've put it on the old (and outdated) port homepage (http://www.dbb.su.se/User:Bjelkmar/Ffcharmm) so that those of you who are interested in this can access and test it. If nothing seems to be wrong after some of you have tested it I'm going to add it to the user contributions on gromacs.org.

Regards,
Pär Bjelkmar

1 okt 2010 kl. 09.53 skrev Sven Jakobtorweihen:

> Dear Pär Bjelkmar,
> 
> thanks for implementing the CHARMM force field into GROMACS, that is really useful. Recently the CHARMM developers have introduced an updated lipid force field CHARMM36 for lipids, see Klauda et al. J. Phys. Chem. B, 2010, 114, 7830 (doi: 10.1021/jp101759q). It seems that this is an important improvement over the preceding version. I checked gromacs version 4.5.1 and the git-repository; unfortunately, the charmm36 lipids are not yet available in gromacs. However, I could write some scripts to convert the charmm files to gromacs files, I have the feeling the bonded parameter conversion is straight forward (having the charmm27 files for gromacs), but I fear that transferring the bonded parameters is tedious. So my question is: Do you have by any change already gromacs files for charmm36 lipids? Or are scripts available to convert charmm FF to gromacs FF files? I am aware of the charmm_to_gromacs perl scripts on the gromacs website. But as fare as I know they are not compatible with the official implementation of charmm in gromacs, or am I wrong?
> 
> Thanks for your help.
> 
> Best regards,
> Sven
> 
> -- 
> *****************************************
> Dr.-Ing. Sven Jakobtorweihen
> 
> Hamburg University of Technology
> Institute of Chemical Reaction Engineering / V2
> Eissendorfer Str. 38
> 21073 Hamburg
> Germany
> 
> PHONE : ++49 (0) 40 42878 2491
> FAX   : ++49 (0) 40 42878 2145
> E-MAIL: jakobtorweihen at tuhh.de
> *****************************************
> 



> 3 sep 2010 kl. 20.56 skrev Drew Bennett:
> 
>> Dear Pr Bjelkmar,
>> 
>> I saw on the gromacs mailing list that you have python a script for converting CHARMM ff to GMX.
>> I would greatly appreciate if you could please send it to me.
>> I am trying to use the new CHARMM36 lipids for a simulation with a membrane protein.
>> 
>> Thank-you.
>> Drew Bennett

> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20101001/b873ef3f/attachment.html>