[gmx-users] (no subject)

[Berk Hess]

>From: Daejoong Kim <daejoongkim at stanford.edu>
>Reply-To: Discussion list for GROMACS users <gmx-users at gromacs.org>
>To: gmx-users at gromacs.org
>Subject: [gmx-users] (no subject)
>Date: Mon, 19 Jul 2004 17:55:24 -0700
>
>Hello all. This is what I posted a couple of weeks ago. Please reply to 
>this.
>
>I have two questions but the first one is very important to me.
>
>1. About LJ parameters: sigma/epsilon vs. c6/c12
>Instead of using "itp" files in GROMACS/share/top, I used the following two
>"itp" files for non-bonded interactions. For me the two files are supposed
>to produce the same results but they turn out to be very different. My
>question is: for switching sigma/epsilon between c6/c12 can I just apply 
>the
>following relations without worrying about the unit issues? (which question
>I thought is right but seemingly wrong now.)
>c6 = 4 * epsilon * sigma^6
>c12 = 4 * epsilon * sigma^12
>
>(1) sigma/epsilon "itp" file
>[ defaults ]
>; func_type   comb-rule   gen-pairs   fudgeLJ     fudgeQQ
>           1           2          no       1.0         1.0

>(2) c6/c12 "itp" file
>[ defaults ]
>; func_type   comb-rule   gen-pairs   fudgeLJ     fudgeQQ
>           1           1          no       1.0         1.0

>
>Well, it is annoying. Besides choosing combination rule in [defaults]
>section, do I need to do different things for two cases? Is it an issue of
>units?
>

If you converted all parameters correctly this should work.
It would be safer to use combination rule 3 for (1) and not 2,
since rules 1 and 3 use the same combination rule (see
page 91 of the manual). But in your case this should not
make a difference, since (I think) you have specified
all combinations explicitly.


>2. About charges and masses
>Charges and masses can be finally changed in [atoms] section under
>[moleculetype], right? I mean those values override the ones specified in
>[atomtypes] or any other places? Am I right?

Yes.

But you probably don't know that you can compare tpr files using:
gmxcheck -s1 ... -s2 ...
This will compare all atom, interactions and parameters between
both files and print any differences.
For these kind of problems this is very useful.

Berk.

_________________________________________________________________
Play online games with your friends with MSN Messenger 
http://messenger.msn.nl/