[gmx-users] Re: gb_saltconc in implicit water simulations

[Per Larsson]

Hi!

Sorry that I missed this discussion, we recently moved the lab and there where quite many things to attend to.

When we did the implementation of implicit solvent into Gromacs, one of the strong points was to see how fast we could make it, to speed up conformational sampling as much as possible. Adding in the extra Debye-Huckel term for a non-zero salt-concentration was not a priority at the time. It might be a bit unfortunate that the option anyway exists in the mdp-file. 
I will try and see if I can find the time to put it in, but that will not happen in the next coming weeks although I agree that it would be very useful to have :-)

When it comes to the use of explicit ions in an implicit solvent, I hate to say that I'm not much of an expert, really. Having said that, and someone else is very welcome to correct me on this point, there does not seem to exist a wide body of literature about this. In addition to the paper that Justin pointed to, there is some work done by Michael Feig (eg. http://www.ncbi.nlm.nih.gov/pubmed/16928023). 
Also I think that particularly with DNA there are problems with the theory itself as it does not handle high salt-concentrations very well.

Cheers
/Per



14 dec 2010 kl. 20.22 skrev devicerandom:

> On 14/12/10 19:04, Justin A. Lemkul wrote:
>>> I was going to send a very similar mail -namely, I have a coarse-grain
>>> system in vacuum where I'd like to simulate a charged polymer. My idea
>>> was that of adding a single "ghost" counter-ion with zero VdW
>>> interactions, perhaps keeping it fixed in one corner of the box, but
>>> I'd like to know how meaningful it is.
>>> 
>> 
>> "In vacuum" and "implicit solvent" are very different concepts. If you
>> are indeed simulating in a vacuum, I would be willing to bet that your
>> charged polymer will rapidly associate with your dummy ion simply due to
>> attractive Coulombic interactions that are not screened by any
>> intervening solvent.
> 
> Heh, I know very well that -and that is one of the reasons I wanted to know how does one cope with that.
> 
> I mean, if it's a single polymer, perhaps one can naively position restrain the ion in a corner of the box, away from the polymer, while the polymer stays on its center of mass. But if I want multiple objects in my box, I guess it's not going to work.
> 
> I'd like very much to use implicit solvent as well but I have to see how deep is the performance impact on my model -after all, one does coarse grain to be raw but *fast* :)
> 
>> 
>> In regards to the original post, I have seen this question asked and
>> ignored several times already, unfortunately. Most of the literature I
>> have read (which primarily uses CHARMM) does not specify how counterions
>> are handled, if at all. Maybe there is some intrinsic compensating
>> factor in CHARMM that I simply don't know about. I found this:
>> 
>> http://dx.doi.org/10.1002/jcc.20874
>> 
>> But again, it uses CHARMM and fancy modified electrostatic models.
>> 
>> I would really hope that someone who knows about the proper GROMACS
>> methodology would comment. This has been a curiosity of mine for some
>> time, as well. I have not had a pressing need to try out the implicit
>> solvent features, but in the absence of being sure of the methodology, I
>> am unlikely to do so.
> 
> Perhaps the developers of the implicit solvent know about that? I hope so...
> 
> m.
> -- 
> gmx-users mailing list    gmx-users at gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/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/Support/Mailing_Lists