[gmx-users] All-atom topologies for lipid head groups

[Justin A. Lemkul]

Ioannis Beis wrote:
> Dear Gromacs users,
> 
> I am a new user and I am trying to study the physical properties of the 
> interactions between cytosolic proteins and lipids. I have created the 
> Berger-Gromos combination, as described by Mr. Justin Lemkul in his 
> KALP15 in DPPC tutorial, as well as Berger-OplsAA, as described by Mr. 
> Chris Neale in gromacs mailing list. All-atom topologies might turn out 
> to be crucial in my work, both in the results and the data analysis 
> (e.g. H-bonds).
> 
> Even if the all-atom model of Opls-AA is used for the proteins, the 
> lipid head groups would still have a united-atom description and 
> therefore the interactions can't be treated in an all-atom level.
> 
> I thought of creating a new topology for the lipid head groups under a 
> head.itp file, by using parameter values from Dundee server and/or 
> Opls-AA, even though I do not know yet exactly how and I feel very 
> unsure about the eventual incorporation into the Berger-OplsAA 
> combination. Then taking single lipids of pre-equilibrated membranes, 
> editing accordingly the coordinate file (i.e. replacement of polar head 
> atoms and use of new coordinates, names and residue names for them) and 
> subsequently building bilayers.
> 
> I would really appreciate if someone could send suggestions. In 
> addition, if someone already has an all-atom description of lipid 
> head-groups compatible with a protein-membrane force field combination 
> with all-atom protein description, I would be grateful if he/she could 
> share.
> 
> Thank you in advance.
> 

It sounds to me like you're trying to create a Frankenstein force field.  Why 
wouldn't a UA representation of the lipids work?  For hydrogen bonding, all 
polar groups are represented with explicit H atoms.  This is generally true of 
all UA force fields, for lipids and proteins alike.

If you really want an AA representation of your molecules, you're probably 
better off just applying the CHARMM force field to the whole system.  The latest 
version of Gromacs supports CHARMM, which includes a number of common lipids.

-Justin

-- 
========================================

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================