[gmx-users] problem with simulation of phosphatidic acid, possibly in my topology?

[maria goranovic]

Dear Mark,

Sorry, perhaps I should have clarified more. I did not just make a blind
substitution of a choline by a H-atom.

First, I did conserve charge.
Second, I did introduce reasonable non-bonded and bonded interaction
parameters for the new H-atom. The non-bonded interaction parameters were
similar to those suggested by the mono-anionic phosphate group on a
threonine or serine.

I guess you will agree force fields are often set up that entire functional
groups from one molecule can be transferred to others. For example, one can
use the same head group for both POPC and DPPC, or conversely, one can use
different types of hydrocarbon chains for the same head group, from 12
carbons to 20 carbons, as has often been done. The only parameters that
might change in getting POPA from POPC are near the H-atom itself, and those
are the ones that I had posted earlier, just to see if anyone could notice a
mistake.

Similar methods have been used in literature earlier, except that a DFT
calculation was made for the mono-anionic phosphate group to get new
charges. The DFT did reveal that charges on DOPA were similar to DOPC.

The main issue I have is that I get a somewhat skewed phosphate group that
leads to unfavorable positions for the oxygen atoms on phosphate. I am
guessing that this is because of the bonded-interactions of the new H-atom.


On Mon, May 5, 2008 at 3:10 PM, Mark Abraham <Mark.Abraham at anu.edu.au>
wrote:

> himanshu khandelia wrote:
>
> > Hi,
> >
> > I am trying to run a simulation of a POPC bilayer mixed with some
> > mono-anionic Phosphatidic acid (POPA) , where the choline group is
> > replaced by a hydrogen atom. However, the energy of my simulation box
> > diverges, and I am trying to fix the problem. I also tried running a
> > simulation of PA solvated in water and a single sodium ion, but even
> > that does not work. I will try to provide as much detail as possible,
> > I hope someone can help point out some obvious error which I have not
> > been able to debug over 3 days. In the simulation of a single PA
> > molecule in water, MDRUN keeps complaining that the distance between
> > atoms H1 and O4 is very high (> 4 nm or so). The problem probably lies
> > in my topology, possibly in the bonded interactions?
> >
>
> Merely replacing a functional group with a hydrogen doesn't make for a
> sensible model physics. First, you probably didn't conserve charge, and
> second the parameters don't necessarily make sense for a particle of mass 1
> rather than mass 12 (plus secondary effect of substituent group), and third
> there's no way to fix both the foregoing and the other issues.
>
> You need at least to find a proper model of POPA and to implement it, or
> if need be, develop one yourself. Search the wiki for Parameterization to
> get a start on the latter, but it is an expert topic.
>
> Mark
>
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at http://www.gromacs.org/search before posting!
> Please don't post (un)subscribe requests to the list. Use the www
> interface or send it to gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>



-- 
Maria G.
Technical University of Denmark
Copenhagen
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20080505/03c5c6d3/attachment.html>