[gmx-users] No default Ryckaert-Bell. types for proline

Wed Jun 3 06:00:36 CEST 2015

Thanks Justin,

I have long been confused by atom names and types in Gromacs topology. I
have just went through the manual again, and it confirms me that what we
read directly in ffbonded.itp under GMXHOME/share/gromacs/top/oplsaa.ff are
atomtypes. In other words, what you have listed in the last email

  C      CT_2   N      CT_3    3     -5.72371 -18.33847  -5.23419  29.29636
0.00000   0.00000 ; Pro (fit to AM1) CD-N-CA-C, JT-R 2/10/97
  C      CT_2   NT     CT_3    3     -5.72371 -18.33847  -5.23419  29.29636
0.00000   0.00000 ; Pro (fit to AM1) CD-N-CA-C, JT-R 2/10/97
  C_3    CT_2   N      CT_3    3     -5.72371 -18.33847  -5.23419  29.29636
0.00000   0.00000 ; Pro COO- terminus.
  C_3    CT_2   NT     CT_3    3     -5.72371 -18.33847  -5.23419  29.29636
0.00000   0.00000 ; Pro COO- terminus.

which was extracted from ffbonded.itp is composed of atomtypes. Meanwhile
what I have stated in the brackets - (CD N CA C), (CD N CA CB) or (CD N CA
HA) are by myself read from the topology file after pdb2gmx. To be exact, I
read

[ dihedrals ]
;  ai    aj    ak    al funct            c0            c1            c2
       c3            c4            c5
   12     1     4     5     3
   12     1     4     6     3
   12     1     4    15     3

from the topology and refer the numbers (e.g. 12 1 4 15) to the above
directive [ atoms ] to find out that it is (CD N CA C). We can also read
from directive [ atoms ] that under the "type" column, it should be
(opls_296 opls_309 opls_285 opls_271). So which is the atomtypes, (CD N CA
C) or (opls_296 opls_309 opls_285 opls_271)? It seems (opls_296 opls_309
opls_285 opls_271) are the atomtypes, but why the itp states  (C CT_2 N
CT_3)?

And how can I refer my topology to this line of parameter when it deals (CD
N CA C) or (opls_296 opls_309 opls_285 opls_271)?

Regards,
Kevin

On Wed, Jun 3, 2015 at 10:11 AM, Justin Lemkul <jalemkul at vt.edu> wrote:

>
>
> On 6/2/15 8:51 PM, Kevin C Chan wrote:
>
>> Thanks Justin,
>>
>> I'm sorry that I'm not clear with what do you mean by "so you can likely
>> just use the same parameters in ffbonded.itp for this dihedral in the
>> zwitterionic form".
>>
>>
> The Pro dihedral term for this torsion is the same, irrespective of
> whether it's N- or C-terminal.  From ffbonded.itp:
>
>   C      CT_2   N      CT_3    3     -5.72371 -18.33847  -5.23419
> 29.29636 0.00000   0.00000 ; Pro (fit to AM1) CD-N-CA-C, JT-R 2/10/97
>   C      CT_2   NT     CT_3    3     -5.72371 -18.33847  -5.23419
> 29.29636 0.00000   0.00000 ; Pro (fit to AM1) CD-N-CA-C, JT-R 2/10/97
>   C_3    CT_2   N      CT_3    3     -5.72371 -18.33847  -5.23419
> 29.29636 0.00000   0.00000 ; Pro COO- terminus.
>   C_3    CT_2   NT     CT_3    3     -5.72371 -18.33847  -5.23419
> 29.29636 0.00000   0.00000 ; Pro COO- terminus.
>
> All the same, so the parameters for the zwitterionic species can likely be
> copied to your missing interaction.
>
>  Also as I have mentioned before, when I look into the line 161, it refers
>> to a dihedral term concerning (CD N CA C) which is not able to be found in
>> ffbonded.itp. However, other dihedral terms such as (CD N CA CB) or (CD N
>> CA HA) which are also not found in itp do not report such error and I was
>>
>
> Atom names and atom types are entirely different concepts.  grompp doesn't
> care what you call your atoms (atom name), but it does care what chemical
> characteristics they have (atom type).
>
>  successful to grompp after commenting out the line 161. So why would this
>> happen?
>>
>>
> Don't ever comment out lines because of issues like this.  By doing so,
> you're saying that torsion contributes nothing to the energy function,
> which is wrong.  I suggest copying the above parameters to the correct
> sequence of atom types.  Stuff like this happens all the time.  Force
> fields are massive entities, and are added to incrementally as need
> arises.  A zwitterionic proline is probably not something that people
> commonly use, and its the most complex case due to the chemical nature of
> it.  It's entirely possible that the .tdb entries were simply added by
> analogy to existing groups as a convenience and never fully tested.
>
>
> -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.
>