AW: about angles in coordination complexes

From: Norman Geist (norman.geist_at_uni-greifswald.de)
Date: Tue Jan 27 2015 - 02:23:29 CST

The force constants seem quite small to me. If they are too small, you will
not even see the effect. Have you actually measured the angles if they are
different from what you set in the extrabonds? maybe you need to restrain
more planes than only the angles to get your conformation. Once you know how
to constrain your conformation you might want to create such a patch for
your FF. You should also think about, based on what interactions this
conformation is actually build in reality, so that you not just force your
conformation in “any way” but create a reasonable model of your system.

 

Norman Geist.

 

Von: luca belmonte [mailto:lucabelmonte_at_gmail.com]
Gesendet: Montag, 26. Januar 2015 18:53
An: Norman Geist
Betreff: Fwd: namd-l: about angles in coordination complexes

 

Dear Norman,

 

I forward you an email in reply to the issue of the external bonds.

Only after I sent it I realized I was missing your address.

 

Thank you,

Best regards,

Luca

 

 

---------- Forwarded message ----------
From: luca belmonte <lucabelmonte_at_gmail.com>
Date: 2015-01-26 18:51 GMT+01:00
Subject: Re: namd-l: about angles in coordination complexes
To: Jérôme Hénin <jerome.henin_at_ibpc.fr>
Cc: Namd Mailing List <namd-l_at_ks.uiuc.edu>

Dear Norman,

Dear Jerome,

 

I attempted using extrabonds as explained here
(http://www.ks.uiuc.edu/Research/namd/2.9/ug/node27.html), but the results
seems to be not affected.

No errors, no warnings.

 

Could you please have a look at the following data? they seems to me
reasonably formatted

angle 1 20 5 1.35337 111.83119 # since atoms are based 0 I decreased of 1
the atom related entries

angle 1 20 7 0.768447 103.15240 # k is in Kcal/rad^2

angle 1 20 3 1.57934 122.83501 # ref in degrees

angle 3 20 5 0.939363 95.91890

angle 3 20 7 1.59095 98.25385

angle 5 20 7 0.832975 126.54048

 

and I recall them using

extraBonds on

extraBondsFile $parFolder/cobalt_extrabonds.txt

 

I'll do meanwhile an attempt with the patches.

 

Thank you for your time,

Best,

Luca

 

2015-01-26 17:55 GMT+01:00 luca belmonte <lucabelmonte_at_gmail.com>:

I am already familiar with patches and I have to admit that I never thought
to them under this perspective.

 

I'll try with the solution proposed by Norman and with the patch.

 

Thank you for the suggestions.

 

 

 

2015-01-26 17:50 GMT+01:00 Jérôme Hénin <jerome.henin_at_ibpc.fr>:

You'll need to create and apply a patch, which is like a special residue
that modifies an existing topology.
http://www.charmm.org/documentation/c34b1/struct.html#%20Patch

You can find an example in the standard CHARMM topology for proteins is you
look for the patch for disulfide bonds (called DISU).

That patch is applied in the following psfgen example:
http://www.ks.uiuc.edu/Research/namd/2.9/ug/node17.html

Jerome

 

 

On 26 January 2015 at 17:28, luca belmonte <lucabelmonte_at_gmail.com> wrote:

Yes, in the QM calculations I have all of them, but I did not mapped them
into the par file.

 

What I have done is to map everything related to the methanethiol group of
the cysteine residue of the top/par files but I did not mapped any angle
between the sulphurs of different cysteines (the S-M2+-S angle you said).

 

Could you please provide me any example, keyword or manual reference to do
that?

 

 

2015-01-26 17:20 GMT+01:00 Jérôme Hénin <jerome.henin_at_ibpc.fr>:

What I said was very generic. Now looking at the specific arrangement, it
should be possible to reliably enforce a tetrahedral shape with 6 S-Mg-S
angle potentials. Do you have those in your model? If yes maybe there's a
problem with them.

Jerome

 

On 26 January 2015 at 17:02, luca belmonte <lucabelmonte_at_gmail.com> wrote:

Hi Jerome,

 

Thank you for the reply.

 

At the very beginning of the story I was wondering to do as you suggest.

But then I encountered some difficulties. Probably I lack the correct
keyword to map the dihedral angle between different residues.

 

Do you have some hint?

 

Thank you,

Luca

 

2015-01-26 16:56 GMT+01:00 Jérôme Hénin <jerome.henin_at_ibpc.fr>:

Hi Luca,

> These complexes were previously parameterized with GAMESS US, results of
its jobs mapped (to the top and par files)

You didn't say what exactly they were mapped to, but your results indicate
that the empirical description you have now is not sufficient to describe
the geometric constraints around your metal center.

 

You probably need more angle/dihedral/improper dihedral terms. I would
include them in the top/par files rather than extraBonds, because they will
be an essential part of your custom force field.

 

Jerome

 

On 26 January 2015 at 16:28, luca belmonte <lucabelmonte_at_gmail.com> wrote:

Thank you for the hint!

 

Best,

Luca

 

2015-01-26 16:24 GMT+01:00 Norman Geist <norman.geist_at_uni-greifswald.de>:

Hi,

 

I think you are less interested in fixing or positional harmonic
restraints, rather than in defining additional bonds/angles/dihedral to
maintain your conformation. this can be done quite easy using the
“ExtraBonds” interface of NAMD. Please find details in the manual.

 

Norman Geist.

 

Von: owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] Im Auftrag
von luca belmonte
Gesendet: Montag, 26. Januar 2015 15:39
An: namd-l_at_ks.uiuc.edu
Betreff: namd-l: about angles in coordination complexes

 

Dear Namd users,

 

I am pretty new to this mail list, so... nice to meet you.

 

Let me introduce briefly the target. I want to simulate several small
complexes made of small peptides. In these peptides cysteines always
coordinate a transition metal dication (M2+). The purpose is to have a guess
of the stability of the complexes by mean of internal energy estimation.

 

The complexes are "rubredoxin" likes, and you can look at coordination
sphere as follows:

 

Cys Cys

       \ /

        M2+

       / \

Cys Cys

 

All the complexes on which I am working on have tetrahedral coordination
shapes, in which the M2+ is in the middle of the cage, while cysteines
sulphurs are at vertexes.

These complexes were previously parameterized with GAMESS US, results of its
jobs mapped (to the top and par files) and till know everything is fine.

 

Now comes the problem. When I look at the shape of the coordination complex,
after the minimization, I expect to see a tetrahedral coordination, but the
result is a non common shape in which the M2+ pop out from the coordination
plane.

 

To address this issue I was wondering to fix the sulphur atoms positions
around the cluster using fixedAtoms, but I am scared that this approach can
affect calculations, providing wrong energy values. Do you think so? Do you
have another strategy to overcome this problem?

 

Any hint is welcome,

Thank you in advance,

 

Best regards,

 

-- 
Luca Belmonte, PhD
 
-- 
Luca Belmonte, PhD
CIBIO - Center for Integrative Biology
University of Trento
 
Phone: +39 0461 28 53 41
Mobile: +39 328 45 81 762
 
 
-- 
Luca Belmonte, PhD
CIBIO - Center for Integrative Biology
University of Trento
 
Phone: +39 0461 28 53 41
Mobile: +39 328 45 81 762
 
 
-- 
Luca Belmonte, PhD
CIBIO - Center for Integrative Biology
University of Trento
 
Phone: +39 0461 28 53 41
Mobile: +39 328 45 81 762
 
 
-- 
Luca Belmonte, PhD
CIBIO - Center for Integrative Biology
University of Trento
 
Phone: +39 0461 28 53 41
Mobile: +39 328 45 81 762
 
-- 
Luca Belmonte, PhD
CIBIO - Center for Integrative Biology
University of Trento
 
Phone: +39 0461 28 53 41
Mobile: +39 328 45 81 762
 
-- 
Luca Belmonte, PhD
CIBIO - Center for Integrative Biology
University of Trento
 
Phone: +39 0461 28 53 41
Mobile: +39 328 45 81 762

This archive was generated by hypermail 2.1.6 : Tue Dec 27 2016 - 23:20:51 CST