[gmx-users] Adapting metallic ions from AMBER to GROMACS

Sat Apr 23 02:49:42 CEST 2016

On 4/22/16 8:40 PM, Pedro Lacerda wrote:
> Applying the same simple conversion to the vdw parameters for atoms Mg+2
> (id 46), Ca+2 (id 47), and Zn+2 (id 48) found in amber99.prm of the TINKER
> package, which is the base of Sorin parameters, there is agreement between
> Mg+2 and Zn+2.
>
> It looks like a bug in the L-J parameters of TINKER package for amber99*
> force fields. If it becomes confirmed, should I fill an issue at Redmine or
> request a code review?
>
> However I may be wrong...
>
> http://dasher.wustl.edu/tinker/distribution/params/amber99.prm
>

You're not looking at the right parameters.  What you've claimed to be Ca2+ from 
ffnonbonded.itp is:

Ca2+ 3.40E-01 3.60E-01

But it's not.  The calcium atom type is C0 (see ions.itp):

C0          20      40.08    0.0000  A   3.05240e-01  1.92376e+00

If you run your conversion on the right parameters you will see it matches exactly.

-Justin

> 2016-04-22 20:11 GMT-03:00 Pedro Lacerda <pslacerda+gmx at gmail.com>:
>
>> For me they must be the same in both force fields amber99sb-ildn, amber99,
>> at the paper, etc. If it is different then should be another force field,
>> isn't it?
>>
>> About the conversion, I did it wrong on my spreadsheet. Sorry...
>>
>> It works! Many thanks!
>>
>> Except for the epsilon of Ca+2. I verified the parameters of atom
>> amber99_15  (Ca +2) in ffamber_v4.0/ffamber99sb/ffamber99sbsbnb.itp and, of
>> course, is the same of GROMACS files, but it is different than the
>> previously referred paper.
>>
>> I also checked the parameters in amber10.ffparms.tar.bz2/parm/parm99.dat,
>> and it agreed with paper, disagreeing with GROMACS.
>>
>> Just a small question, PARM is the file format for force fields in AMBER,
>> right?
>>
>> http://ffamber.cnsm.csulb.edu/ffamber.php
>> http://ambermd.org/dbase.html
>>
>>
>>
>> 2016-04-22 19:29 GMT-03:00 Justin Lemkul <jalemkul at vt.edu>:
>>
>>>
>>>
>>> On 4/22/16 6:24 PM, Pedro Lacerda wrote:
>>>
>>>> Thank you!
>>>>
>>>> According to [1, Table 2a], these are the L-J (6-12) parameters for Ca2+
>>>> used at AMBER package. Or the paper or GROMACS copied wrongly from the
>>>> AMBER package source because simple unit conversion gives different
>>>> results.
>>>>
>>>>
>>> That paper was published 3 years after AMBER99SB-ILDN so expecting those
>>> parameters to be in that force field implementation is wrong.  There could
>>> be a mistake but the AMBER ports have been extensively verified.  There are
>>> lots of AMBER-compatible ion parameters floating around, though.
>>>
>>> I don't own a copy of AMBER package, please could you verify the Ca L-J
>>>> parameters for me?
>>>>
>>>>
>>> The AMBER force fields should be freely available (I don't use AMBER
>>> software).
>>>
>>> And I can't see how your conversion formula from Rmin/2 to sigma could
>>>> produce the right result.
>>>>
>>>>
>>> Plug in the values, it works.  It's the same conversion for CHARMM.  For
>>> instance, Mg2+:
>>>
>>> 1.1 * 2 * 2^(-1/6) * 0.1 = 0.196
>>>
>>> -Justin
>>>
>>>
>>> [1] J Chem Theory Comput. 2013 June 11; 9(6): 2733–2748.
>>>> doi:10.1021/ct400146w.
>>>>
>>>> 2016-04-21 23:35 GMT-03:00 Justin Lemkul <jalemkul at vt.edu>:
>>>>
>>>>
>>>>>
>>>>> On 4/21/16 4:30 PM, Pedro Lacerda wrote:
>>>>>
>>>>> Hi Mark,
>>>>>>
>>>>>> I starting converting from AMBER (according to the paper) to GROMACS
>>>>>> and
>>>>>> comparing the results.
>>>>>>
>>>>>> Please explain in more words what this constant of two mean or how can
>>>>>> I
>>>>>> handle it. Ignoring it and just converting the units, I got these
>>>>>> results
>>>>>>
>>>>>>
>>>>>> It's force field convention, just different notation.  Sometimes
>>>>> different
>>>>> force fields or programs use Rmin/2, Rmin, or sigma.  They're easily
>>>>> converted, e.g.:
>>>>>
>>>>> sigma = (Rmin/2)*2*2^(-1/6)
>>>>>
>>>>> and in GROMACS the units are nm so you need an additional multiplication
>>>>> by 0.1 to convert A -> nm.
>>>>>
>>>>> AMBER (table 2a)
>>>>>
>>>>>> Rmin/2 (Å) epsilon (kcal/mol)
>>>>>> Zn2+ 1.10E+00 1.25E-02
>>>>>> Mg2+ 7.93E-01 8.95E-01
>>>>>> Ca2+ 1.71E+00 4.60E-01
>>>>>> AMBER converted
>>>>>> R (nm) epsilon (kJ/mol)
>>>>>> Zn2+ 2.20E+01 5.23E-02
>>>>>> Mg2+ 1.59E+01 3.74E+00
>>>>>> Ca2+ 3.43E+01 1.92E+00
>>>>>>
>>>>>> amber99sb-ildn.ff/ffnonbonded.itp
>>>>>> sigma (nm) epsilon (kJ/mol)
>>>>>> Zn2+ 1.96E-01 5.23E-02
>>>>>> Mg2+ 1.41E-01 3.74E+00
>>>>>> Ca2+ 3.40E-01 3.60E-01
>>>>>>
>>>>>> I got exact matches for epsilon in Zn and Mg, but not for Ca.  Also the
>>>>>> difference on sigma was in the order Å/10, except for Zn which was
>>>>>> relatively significant.
>>>>>>
>>>>>>
>>>>>> See above for the proper way to convert Rmin/2 to sigma.
>>>>>
>>>>> As for Ca, probably the parameters are coming from a different source.
>>>>> I
>>>>> don't know what "Table 2a" means or what paper it's from, but you need
>>>>> to
>>>>> verify that the source of parameters is compatible or that you
>>>>> understand
>>>>> where differences are coming from.
>>>>>
>>>>> -Justin
>>>>>
>>>>> --
>>>>> ==================================================
>>>>>
>>>>> Justin A. Lemkul, Ph.D.
>>>>> Ruth L. Kirschstein NRSA Postdoctoral Fellow
>>>>>
>>>>> Department of Pharmaceutical Sciences
>>>>> School of Pharmacy
>>>>> Health Sciences Facility II, Room 629
>>>>> University of Maryland, Baltimore
>>>>> 20 Penn St.
>>>>> Baltimore, MD 21201
>>>>>
>>>>> jalemkul at outerbanks.umaryland.edu | (410) 706-7441
>>>>> http://mackerell.umaryland.edu/~jalemkul
>>>>>
>>>>> ==================================================
>>>>>
>>>>> --
>>>>> 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.
>>>>>
>>>>>
>>> --
>>> ==================================================
>>>
>>> Justin A. Lemkul, Ph.D.
>>> Ruth L. Kirschstein NRSA Postdoctoral Fellow
>>>
>>> Department of Pharmaceutical Sciences
>>> School of Pharmacy
>>> Health Sciences Facility II, Room 629
>>> University of Maryland, Baltimore
>>> 20 Penn St.
>>> Baltimore, MD 21201
>>>
>>> jalemkul at outerbanks.umaryland.edu | (410) 706-7441
>>> http://mackerell.umaryland.edu/~jalemkul
>>>
>>> ==================================================
>>> --
>>> 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.
>>>
>>
>>

-- 
==================================================

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==================================================