[gmx-users] Conversion of the GLYCAM dihedral parameters in the GROMACS format.

[Mark Abraham]

On 20/07/2011 7:33 PM, sa wrote:
> Thank you Austin, for the clarification,
>
> So my questions are: how to convert it in the gromacs format ? Is it 
> correct to use the absolute value of the multiplicity in my parameters 
> and use the negative values of barrier heights when they are exist in 
> the AMBER parameters? Since, I have noticed that in the AMBER ff port 
> in GROMACS, the multiplicity values are always set to > 0 and  the 
> barrier heights have sometime a negative value.

You want to implement the same mathematical function in each case. Take 
the GLYCAM parameters, drop them into the correct function. Then consult 
chapters 4 and 5 of the GROMACS manual and see what will be necessary to 
reproduce that function. Then test on a tiny system that you can also 
compute by hand to confirm your understanding.

Mark

>
> Thank you again for your advice.
>
> SA-
>
>
>
>     Message: 1
>     Date: Tue, 19 Jul 2011 11:08:55 -0700 (PDT)
>     From: "Austin B. Yongye" <ybausty at yahoo.com
>     <mailto:ybausty at yahoo.com>>
>     Subject: Re: [gmx-users] Conversion of the GLYCAM dihedral parameters
>            in the  GROMACS format.
>     To: Discussion list for GROMACS users <gmx-users at gromacs.org
>     <mailto:gmx-users at gromacs.org>>
>     Message-ID:
>     <1311098935.93905.YahooMailClassic at web161426.mail.bf1.yahoo.com
>     <mailto:1311098935.93905.YahooMailClassic at web161426.mail.bf1.yahoo.com>>
>     Content-Type: text/plain; charset="utf-8"
>
>     It is normal to have combinations of negative and positive values
>     for the barrier heights. Those are just the best coefficients to
>     reproduce some QM rotational energy curve during the
>     parameterization.  The negative periodicities are a convention
>     from AMBER. They simply indicate that the dihedral angle potential
>     has more than one term. For your example below:
>
>     O2-P
>              -OS-CP   1    0.10          0.0            -3.        
>     Dimethyl phosphate
>                                 1   -0.50          0.0            -2.
>
>              1    0.10
>              0.0             1
>     -3, -2 and 1 signify V3, V2 and V1 terms, respectively. Once a
>     positive value is reached, terms for the O2-P-OS-CP potential have
>     been completely accounted for.
>
>     Hope that helps.
>     Austin-
>
>
>     --- On Tue, 7/19/11, Mark Abraham <Mark.Abraham at anu.edu.au
>     <mailto:Mark.Abraham at anu.edu.au>> wrote:
>
>     From: Mark Abraham <Mark.Abraham at anu.edu.au
>     <mailto:Mark.Abraham at anu.edu.au>>
>     Subject: Re: [gmx-users] Conversion of the GLYCAM dihedral
>     parameters in the GROMACS format.
>     To: "Discussion list for GROMACS users" <gmx-users at gromacs.org
>     <mailto:gmx-users at gromacs.org>>
>     Date: Tuesday, July 19, 2011, 7:08 AM
>
>
>
>
>
>
>
>        On 19/07/2011 11:56 PM, sa wrote:
>
>
>
>
>
>
>
>
>
>
>          Dear
>              GROMCS users,
>
>          I
>              am trying to convert some GLYCAM parameters in GROMACS
>     format.
>              For this
>              purpose, I am using the latest GLYCAM parameters downloaded
>              from the RJ. Woods'
>              website and the examples given in the acpype code (here for
>              the dihedral angles) :
>
>
>
>
>
>     http://code.google.com/p/acpype/source/browse/trunk/acpypi.py?spec=svn9&r=9&redir=1
>     <http://code.google.com/p/acpype/source/browse/trunk/acpypi.py?spec=svn9&r=9&redir=1>
>
>
>
>
>
>          -------------------
>     # dihedral    idivf        barrier hight/2 kcal/mol  phase degrees
>           periodicity     comments
>           X -ca-ca-X    4           14.500*                    
>     180.000                2.000             intrpol.bsd.on C6H6
>
>     * to convert to gmx: 14.5/4 * 4.184 * 2 (?) (yes in amb2gmx, no in
>     topolbuild, why?) = 30.334 or 15.167 kJ/mol
>     # X -CA-CA-X    4   14.50        180.0             2.        
>     intrpol.bsd.on C6H6 (from parm99.dat)
>
>     # X-CA-CA-X     3    30.33400     0.00000   -30.33400     0.00000
>         0.00000     0.00000   ; intrpol.bsd.on C6H6
>
>     -----------
>
>          I have no problems with the parameters for proteins. But, in
>     case of the GLYCAM parameters, I am a little confused
>
>     about the conversion of dihedral force constants (DFC), especially
>     when the DFC and the periodicity values are < 0 for example
>     for this torsion:
>
>
>          O2-P
>              -OS-CP   1    0.10          0.0            -3.        
>     Dimethyl phosphate
>                        1   -0.50          0.0            -2.
>
>              1    0.10
>              0.0             1
>
>
>
>        Where only a positive value makes sense, sometimes people use
>        negative values to indicate some special functional form. This can
>        be easier to code. Regardless, you'll have to check out the GLYCAM
>        documentation and see what is meant, before you can address how to
>        convert it into a GROMACS format. Obviously the contents of
>     parts of
>        chapter 4 and 5 of the manual will be important.
>
>
>
>        Mark
>
>
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20110721/de145afc/attachment.html>