Re: [AMBER] LJEDIT syntax

From: Matias Machado <mmachado.pasteur.edu.uy>
Date: Tue, 24 Jul 2018 17:58:15 -0300 (UYT)

Thanks David,

With your comments I was able to implement the special LJ parameters into the frcmod file.

However, I notice that the actual parameters read for each atom type are "Rm/2" (= Rmh, "Rm half") and "epsilon" (e), same as in the NONBON section.

So sigma_ij = (Rmh[type1] + Rmh[type2])*2^(1/6) and e_ij = sqrt(e[type1] * e[type2])

The other important thing is that the section LJEDIT only affects the defined interaction (it doesn't overwrite the NONBON section). Indeed these three definitions of the A-B interaction render the same prmtop despite being different from a conceptual point of view:

LJEDIT
# type1 type1 Rmh[type1] e[type1] Rmh[type2] e[type2]
  A B 1.5 0.2 1.5 0.2
  A B 3.0 0.2 0.0 0.2
  A B 1.0 0.1 2.0 0.4

By the way, I think this is a very nice feature that introduces a lot of flexibility to AMBER (comparable to other popular MD engines), I hope it won't be deprecated...

Best,

Matías

------------------------------------
PhD.
Researcher at Biomolecular Simulations Lab.
Institut Pasteur de Montevideo | Uruguay
[http://pasteur.uy/en/laboratorios-eng/lsbm]
[http://www.sirahff.com]

----- Mensaje original -----
De: "David Cerutti" <dscerutti.gmail.com>
Para: "AMBER Mailing List" <amber.ambermd.org>
Enviados: Lunes, 23 de Julio 2018 16:09:38
Asunto: Re: [AMBER] LJEDIT syntax

I'm the culprit here. I came up with that when we thought we needed
parameters off the usual combining matrix. What the numbers stand for is
the Lennard-Jones sigma and epsilon of the first atom type when it
interacts with anything under the normal rules, then the sigma and epsilon
of the second atom type when it interacts with the first. So all of those
OW's in the parm14ipq.dat file have the sigma and epsilon of TIP4P-Ew water
oxygen (the OW parameters are recast from the TIP3P which I sloppily left
in the list above--TIP4P-Ew is the water model of choice for ff14ipq).
Then, when type ND interacts with OW, it does so as if the ND has sigma and
epsilon 2^(5/6) * 1.960000 A and 0.170000 kcal/mol. So the interaction of
ND and OW takes place according to

sigma = 0.5 * (1.775931 + 1.960000) * 2^(5/6) = 3.3283A
epsilon = sqrt(0.162750 * 0.170000) = 0.16634 kcal/mol

That's the way LJEDIT works. I'll see about getting this into the manual,
but it's rare that someone has wanted the feature.

Dave



On Mon, Jul 23, 2018 at 2:16 PM Matias Machado <mmachado.pasteur.edu.uy>
wrote:

> Dear AMBER experts,
>
> I'm trying to set some Lennard-Jones interactions out-off the combination
> rule. I was able to do it by using Parmed, however I want to set them in a
> frcmod file so they can be applied by leap without further modification of
> the topology.
>
> I know the LJEDIT section (available from amber14 on) would allow me to do
> so, however its syntax is poorly documented, indeed I couldn't find any
> mention to that keyword in the Amber manuals, nor in the file formats at
> the AMBER web page [http://ambermd.org/FileFormats.php]
>
> The only reference I found was this post in the mailing list [
> http://dev-archive.ambermd.org/201301/0018.html], in which its
> implementation is discussed. From those comments, I would expect the
> following syntax:
>
> LJEDIT
> <Type 1> <Type 2> <R* pair> <Eps pair>
> ...
> END
>
> However, looking at the implementation of parm14ipq.dat [
> https://www.ncbi.nlm.nih.gov/pubmed/25328495] and CUFIX
> (frcmod.ff99cufix) [http://bionano.physics.illinois.edu/CUFIX]
> I found 4 terms instead of 2 for defining the LJ interaction, my guess is
> that 1-4 interactions can also be set separately but I'm not sure about
> that...
>
> In any case... what does each set of parameters stand for?
>
> Best regards,
>
> Matías
>
> ------------------------------------
> PhD.
> Researcher at Biomolecular Simulations Lab.
> Institut Pasteur de Montevideo | Uruguay
> [http://pasteur.uy/en/laboratorios-eng/lsbm]
> [http://www.sirahff.com]
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber

_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Jul 24 2018 - 14:00:06 PDT
Custom Search