[gmx-developers] problems with parameter assignement to CSH and COH groups when "-vsite hydrogen" flag is used
Karmen Condic-Jurkic
k.condicjurkic at uq.edu.au
Thu Jan 16 11:01:14 CET 2014
Hi Berk,
Thanks for the swift reply. I looked at the thread on Redmine and tried to use constraint=h-angle with -vsite topology of my test system, but grompp froze when it started to turn all bonds and H-angles to constraints. But the reason could easily be my lack of experience with Gromacs...
I just wanted to emphasize that the topologies generated by pdb2gmx with "-vsite h" are able to correctly pick up the atoms of interest and list them in the constraint block, but it's just that grompp is unable to differentiate between COH and CSH constraints. It fails to create a specific function type for CSH because it doesn't see it as a different constraint compared to the COH due to the definition via atom types. I must admit I don't see why calculating the parameters directly from the bond distances and angles would negatively affect other topologies, especially because these constraints are defined in a separate list from others, if I interpreted the code correctly. Another way that comes to my mind is checking the middle atom type when constraint is between n-th and n+2-th neighbour, but that is not necessarily simpler solution.
Cheers,
Karmen
-----------------------------
Dr. Karmen Condic-Jurkic
Postdoctoral Researcher
School of Chemistry and Molecular Biosciences
The University of Queensland
Brisbane St. Lucia, QLD 4072
Australia
________________________________________
From: gromacs.org_gmx-developers-bounces at maillist.sys.kth.se [gromacs.org_gmx-developers-bounces at maillist.sys.kth.se] on behalf of gromacs.org_gmx-developers-request at maillist.sys.kth.se [gromacs.org_gmx-developers-request at maillist.sys.kth.se]
Sent: Wednesday, January 15, 2014 7:19 PM
To: gromacs.org_gmx-developers at maillist.sys.kth.se
Subject: gromacs.org_gmx-developers Digest, Vol 117, Issue 14
Send gromacs.org_gmx-developers mailing list submissions to
gromacs.org_gmx-developers at maillist.sys.kth.se
To subscribe or unsubscribe via the World Wide Web, visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-developers
or, via email, send a message with subject or body 'help' to
gromacs.org_gmx-developers-request at maillist.sys.kth.se
You can reach the person managing the list at
gromacs.org_gmx-developers-owner at maillist.sys.kth.se
When replying, please edit your Subject line so it is more specific
than "Re: Contents of gromacs.org_gmx-developers digest..."
Today's Topics:
1. problems with parameter assignement to CSH and COH groups
when "-vsite hydrogen" flag is used (Karmen Condic-Jurkic)
2. Re: problems with parameter assignement to CSH and COH groups
when "-vsite hydrogen" flag is used (Berk Hess)
----------------------------------------------------------------------
Message: 1
Date: Wed, 15 Jan 2014 05:55:28 +0000
From: Karmen Condic-Jurkic <k.condicjurkic at uq.edu.au>
To: "gromacs.org_gmx-developers at maillist.sys.kth.se"
<gromacs.org_gmx-developers at maillist.sys.kth.se>
Subject: [gmx-developers] problems with parameter assignement to CSH
and COH groups when "-vsite hydrogen" flag is used
Message-ID:
<362AA6E0D8FA7244B898E0BFA9ECD906B0DC7D at uqexmdb4.soe.uq.edu.au>
Content-Type: text/plain; charset="iso-8859-1"
Hi everyone,
I'm a new Gromacs user and I bumped into a problem with the parameter assignment to the constraint bonds in COH and CSH groups in serines and cysteines when flag -vsite hydrogens is being used in the topology preparation. Namely, after running an energy minimisation on the system built with virtual sites, I noticed that CSH angles in cysteines ended up with a strange pointy form corresponding to 75 degrees, instead of the equilibrium value of 96. After a bit of a digging, it turned out that the problem is created during the grompp run, where only a single function is created to represent the constraint bond that prevents angle bending for both COH and CSH groups. There parameter used in that particular function corresponds to the COH group (0.198 nm), which is too short for CSH (should be 0.237 nm) and results with the distorted CSH angle. The reason why this happens seems to be the vague definition of the constraint bond given through the atom types (CH2 and H), which are un
fortunately the same for both CYS and SER. In addition, the CH2-H distance parameter for CSH is actually completely omitted from the current ffG54a7bon.file! There is no problem with other COH groups, which are defined through different atom types, as shown below in the excerpt from ffG54a7bon.itp file with parts of interest highlighted:
*************************************************************************
; get the constraint distances for dummy atom constructions
#include "ff_dum.itp"
[ constrainttypes ]
; now the constraints for the rigid NH3 groups
MNH3 C 2 DC_MNC1
MNH3 CH1 2 DC_MNC2
MNH3 CH2 2 DC_MNC2
MNH3 MNH3 2 DC_MNMN
; and the angle-constraints for OH and SH groups in proteins:
CH2 H 2 DC_CO
CH1 H 2 DC_CO
C H 2 DC_CO
P H 2 DC_PO
????????????????????? missing DC_CS value
**************************************************************************
I'm not really sure if this belongs to the developers or to the users mailing list, but I gues that depends on the solution of the problem. It can be probably done on a more general level by redefining the constraints using more unique atom descriptors or just by calculating the atom distance for each constraint of interest by looking at the atoms directly and not reading in a fixed value. As I have no experience in coding in C, this goes way beyond my programming skills so I came up with an easier solution - introducing additional atom type for hydrogen bound to sulphur named HS. With this, the grompp was able to override this issue and this seems to be working:
**************************************************************************
; get the constraint distances for dummy atom constructions
#include "ff_dum.itp"
[ constrainttypes ]
; now the constraints for the rigid NH3 groups
MNH3 C 2 DC_MNC1
MNH3 CH1 2 DC_MNC2
MNH3 CH2 2 DC_MNC2
MNH3 MNH3 2 DC_MNMN
; and the angle-constraints for OH and SH groups in proteins:
CH2 H 2 DC_CO
CH1 H 2 DC_CO
C H 2 DC_CO
P H 2 DC_PO
CH2 HS 2 DC_CS
************************************************************************
I duplicated all the parameters given for atom type H and reassigned them to HS type. The modified force field files (version G54a7) are in the attachment and have suffix _HS.
I was trying to find if this problem was already known, but I couldn't find anything related to this issue. In any case, it would be good to fix this as many proteins contain cysteines, but in case it has already been fixed, I apologize for spamming.
All the best,
Karmen
-----------------------------
Dr. Karmen Condic-Jurkic
Postdoctoral Researcher
School of Chemistry and Molecular Biosciences
The University of Queensland
Brisbane St. Lucia, QLD 4072
Australia
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20140115/fbea5b5b/attachment-0001.html>
------------------------------
Message: 2
Date: Wed, 15 Jan 2014 10:19:30 +0100
From: Berk Hess <hess at kth.se>
To: gmx-developers at gromacs.org
Subject: Re: [gmx-developers] problems with parameter assignement to
CSH and COH groups when "-vsite hydrogen" flag is used
Message-ID: <52D652A2.1020507 at kth.se>
Content-Type: text/plain; charset="iso-8859-1"; Format="flowed"
Hi,
i think this is an unknown issue. It probably only shows up with the
GROMOS force field, since all other force fields supplied with Gromacs
have an HS atom type. It would indeed be better to automatically
determine the constraint length for the bond lengths and the angle. But
I don't know if we can do this, since this would also auto-generate
parameters for other topologies where, maybe, this is not desirable.
I created a redmine issue:
http://redmine.gromacs.org/issues/1422
Cheers,
Berk
On 01/15/2014 06:55 AM, Karmen Condic-Jurkic wrote:
> Hi everyone,
>
>
> I'm a new Gromacs user and I bumped into a problem with the parameter
> assignment to the constraint bonds in COH and CSH groups in serines
> and cysteines when flag /-vsite hydrogens/ is being used in the
> topology preparation. Namely, after running an energy minimisation on
> the system built with virtual sites, I noticed that CSH angles in
> cysteines ended up with a strange pointy form corresponding to 75
> degrees, instead of the equilibrium value of 96. After a bit of a
> digging, it turned out that the problem is created during the grompp
> run, where only a single function is created to represent the
> constraint bond that prevents angle bending for both COH and CSH
> groups. There parameter used in that particular function corresponds
> to the COH group (0.198 nm), which is too short for CSH (should be
> 0.237 nm) and results with the distorted CSH angle. The reason why
> this happens seems to be the vague definition of the constraint bond
> given through the atom types (CH2 and H), which are unfortunately the
> same for both CYS and SER. In addition, the CH2-H distance parameter
> for CSH is actually completely omitted from the current
> ffG54a7bon.file! There is no problem with other COH groups, which are
> defined through different atom types, as shown below in the excerpt
> from ffG54a7bon.itp file with parts of interest highlighted:
>
> *************************************************************************
> ; get the constraint distances for dummy atom constructions
>
> #include "ff_dum.itp"
>
> [ constrainttypes ]
> ; now the constraints for the rigid NH3 groups
> MNH3 C 2 DC_MNC1
> MNH3 CH1 2 DC_MNC2
> MNH3 CH2 2 DC_MNC2
> MNH3 MNH3 2 DC_MNMN
> ; and the angle-constraints for OH and SH groups in proteins:
> CH2 H 2 DC_CO
> CH1 H 2 DC_CO
> C H 2 DC_CO
> P H 2 DC_PO
> ????????????????????? missing DC_CS value
> **************************************************************************
>
> I'm not really sure if this belongs to the developers or to the users
> mailing list, but I gues that depends on the solution of the problem.
> It can be probably done on a more general level by redefining the
> constraints using more unique atom descriptors or just by calculating
> the atom distance for each constraint of interest by looking at the
> atoms directly and not reading in a fixed value. As I have no
> experience in coding in C, this goes way beyond my programming skills
> so I came up with an easier solution - introducing additional atom
> type for hydrogen bound to sulphur named HS. With this, the grompp was
> able to override this issue and this seems to be working:
>
> **************************************************************************
> ; get the constraint distances for dummy atom constructions
>
> #include "ff_dum.itp"
>
> [ constrainttypes ]
> ; now the constraints for the rigid NH3 groups
> MNH3 C 2 DC_MNC1
> MNH3 CH1 2 DC_MNC2
> MNH3 CH2 2 DC_MNC2
> MNH3 MNH3 2 DC_MNMN
> ; and the angle-constraints for OH and SH groups in proteins:
> CH2 H 2 DC_CO
> CH1 H 2 DC_CO
> C H 2 DC_CO
> P H 2 DC_PO
> CH2 HS 2 DC_CS
>
> ************************************************************************
>
>
> I duplicated all the parameters given for atom type H and reassigned
> them to HS type. The modified force field files (version G54a7) are in
> the attachment and have suffix _HS.
>
> I was trying to find if this problem was already known, but I couldn't
> find anything related to this issue. In any case, it would be good to
> fix this as many proteins contain cysteines, but in case it has
> already been fixed, I apologize for spamming.
>
> All the best,
>
> Karmen
>
>
>
> -----------------------------
> Dr. Karmen Condic-Jurkic
> Postdoctoral Researcher
> School of Chemistry and Molecular Biosciences
> The University of Queensland
> BrisbaneSt. Lucia, QLD 4072
> Australia
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20140115/3309f1a5/attachment.html>
------------------------------
--
Gromacs Developers mailing list
* Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_List before posting!
* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-developers or send a mail to gmx-developers-request at gromacs.org.
End of gromacs.org_gmx-developers Digest, Vol 117, Issue 14
***********************************************************
More information about the gromacs.org_gmx-developers
mailing list