[gmx-users] AMBER nonbond params in GROMACS

[Patrick Charchar]

Hi Mark,

Thank you for your clear and thorough response! I can see how I completely
missed that.

Makes perfect sense now. :)

Kind regards,
Patrick

On 2 February 2016 at 19:59, Mark Abraham <mark.j.abraham at gmail.com> wrote:

> Hi,
>
> Thanks for the attention to detail! :-)
>
> On Tue, Feb 2, 2016 at 6:07 AM Patrick Charchar <
> patrick.charchar at rmit.edu.au> wrote:
>
> > Hello Gromacs Users,
> >
> > Sorry if this is trivial, but can someone please explain to me why LJ
> sigma
> > values in the gromacs implemented amber FFs (ffnonbonded.itp) are
> > multiplied by an additional factor of two when converting them from the
> > values given in the amber papers?
> >
>
> The AMBER paper you cite isn't very clear on this point, but its parameter
> is called R*, and not Rmin. Both AMBER and CHARMM express such parameters
> in force field files as R* = Rmin/2, because that is convenient for
> implementing the combination rule Rmin_ij = (Rmin_i + Rmin_j)/2. See e.g.
> subscript j to Table 14 of the paper you mentioned. (One doesn't want to do
> the division in the inner force loops, but it would have been nice to
> implement that optimization after reading the Rmin parameter, rather than
> in the value that the user might read or write.) You can see e.g. that the
> acpype conversion tool has the same factor of 2 you found you needed
>
> http://svn.code.sf.net/p/ccpn/code/branches/stable/ccpn/python/acpype/acpype.py
> .
>
> GROMACS implements multiple kinds of combination rules, so has to work in a
> more general way, and its dedicated grompp pre-processor helps keep such
> implementation details where they belong. :-)
>
> Mark
>
> To check my sanity... the van der Waals parameters in amber are given as
> > R_min (in angstrom) and epsilon (in kcal/mol) - see Cornell, W.D., et
> al.,
> > J. Am. Chem. Soc. 117 (19), 5179-5197 (1995).
> >
> > So to convert to gromacs units:
> > sigma (in nm) = R_min (in angstrom) / 10 * 2^(-1/6)
> > epsilon (in kJ/mol) = epsilon (in kcal/mol) * 4.184
> >
> >
> > Using sp3 carbon as an example, from amber:
> > R_min = 1.9080 angstrom
> > epsilon = 0.1094 kcal/mol
> >
> > After converting I get:
> > sigma = 0.169983475 nm
> > epsilon = 0.4577296 kJ/mol
> >
> > whereas in ffnonbonded.itp:
> >
> > [ atomtypes ]
> > ; name      at.num  mass     charge ptype  sigma      epsilon
> > CT           6      12.01    0.0000  A   3.39967e-01  4.57730e-01
> > Have I missed something?
> >
> > Thank you for your time.
> > Regards,
> >
> > --
> > Mr. Patrick Charchar
> > PhD Candidate
> >
> > School of Aerospace, Mechanical and Manufacturing Engineering
> > RMIT University
> > Melbourne Victoria 3001,
> > Australia
> >
> > Office phone: +(61 3) 9925 3239
> > http://www.rmit.edu.au/staff/patrick-charchar
> > --
> > Gromacs Users mailing list
> >
> > * Please search the archive at
> > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_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-users or
> > send a mail to gmx-users-request at gromacs.org.
> >
> --
> Gromacs Users mailing list
>
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_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-users or
> send a mail to gmx-users-request at gromacs.org.
>



-- 
Mr. Patrick Charchar
PhD Candidate

School of Aerospace, Mechanical and Manufacturing Engineering
RMIT University
Melbourne Victoria 3001,
Australia

Office phone: +(61 3) 9925 3239
*http://www.rmit.edu.au/staff/patrick-charchar
<http://www.rmit.edu.au/staff/patrick-charchar>*