> On Feb 17, 2015, at 4:42 PM, Marek Maly <marek.maly.ujep.cz> wrote:
>
> Here
>
> ----
>> dielc should be 1, the perl version is wrong here
> ----
>
> you are probably wrong. As far as I remember, from the very start of my
> work with
> Amber (2008) there was "DIELC 4" as the default in the mm_pbsa.in
> example files for NON-GB calculation.
I admit that I misread your previous email. The dielc=4 case was specifically when the GB model was set to 0. If dielc=4 when igb=1, then that *is* clearly wrong.
> I can demonstrate that value 4 is more reasonable than 1 directly on my
> test calculation:
>
> python version (nmode_igb=0)
> DELTA S total= -100.1086 0.0000
> 0.0000
>
> perl version ("IGB 0","DIELC 4" )
> TSTOT -32.26 0.00
I don’t agree with the premise of your demonstration (i.e., that your desired target here is the GB entropy), but this is more philosophical than not.
> perl version is here much closer to the results which uses GB (cca -40
> python, cca -20 perl).
>
> Also if dielc is multiplicative scaling parameter seems to me not so
> useful to have this constant equal 1 by default.
I don’t follow. Just because a scaling parameter exists doesn’t mean it should be used...?
> So my opinion is opposite here i.e. I would say that dielc should be 4 by
> default so that value in
> python version (1) is wrong and should be changed to 4.
I disagree here, and the Python default almost certainly won't change. There are two use cases I see:
1. You want to model aqueous solute entropies with gb=1, dielc=1
2. You want to model gas phase entropies with gb=0, dielc=1
Neither of those use-cases use anything except dielc=1. And gb=1 is better than dielc=4 (or any other constant dielectric or distance-based dielectric) for aqueous solute entropies IMO.
> ------------
> (this should never be !=1 for GB). From the NAB user’s manual describing
> the “dielc” variable:
> -------------
>
> But sorry if GB is used dielc parameter is not used in calculation (so
> should be ignored/skipped in the code) so it's actual value in case of
> calc. with nmode_igb=IGB=1 is irrelevant or am I wrong ?
dielc is not ignored by nab if gb=1. But based on rereading your previous email, it seems like dielc is *not* set to 4 when gb=1 in mm_pbsa.pl (this was my mistake -- I thought that it was).
HTH,
Jason
--
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Feb 17 2015 - 16:30:03 PST
