[gmx-users] IMPLICIT SOLVENT IN AMBER

[Michael Brunsteiner]

I recently came across a paper (J. Phys. Chem. B 2005, 109,
10441-10448) in which  the authors found that using implicit
solvent with explicit ions you easily end up with all the
ions cuddling up in clusters ... which is obviously an artefact.

That is, if you do use this combination you certainly need to
carefully choose your parameters and system set-up, and
run thorough tests before doing any production runs.

... my five cents
mic




Per Larsson wrote:

Hi!

The answer is that I do not now critical it is.
I have seen some papers that seem to hint at it not being overly critical, but, 
again, life does not come at a net charge.
You'd have to see for yourself in your system, given the observables that are 
important I suppose.
Possibly we can implement this in the future, as there indeed seems to be some 
interest in it, but that is not a task I can deal with currently.

/Per


10 mar 2011 kl. 10.52 skrev Yulian Gavrilov:

> Thanks again!
> Don't you know how to make a total charge = 0 in this case, if implicit salt 
>concentration is not implemented currently? Or it is not critically? 
>
> 
> 2011/3/10 Per Larsson <per.larsson at sbc.su.se>
> Hi!
> 
> Yes, except that in point 2, I'm not sure about the effects of explicit ions in 
>an implicit solvent. 
>
> Do deal with that properly one should use an implicit salt concentration, but 
>that is not implemented currently. 
>
> The choice of water-model with pdb2gmx is not important. You can choose 'None' 
>here.