Dear David,
Thanks a lot for your help! RESP charges + QM optimization to pick up
the best available gaff parameters could be indeed the optimal way to do it.
Nevertheless, it seems I found the way to proceed with AM1BCC almost
automatically: I tried to carry out an optimization of geometry and
charge calculations in Chimera using gaff. Only one atomic type was
recognized wrongly, while the ring atoms were correctly assigned, and
the charges on two rings were very similar. So, as you suggested, I
changed this atomic type manually. My first impression, that the
molecule behaves well in the MD. However, just to be sure, a RESP check
should be nice to make in addition.
Thank you and cheers,
Sergey
On 16.08.2018 17:46, David A Case wrote:
> On Thu, Aug 16, 2018, Sergey Samsonov wrote:
>> I'm trying to get gaff parameters for a molecule of berenil following the
>> standard procedure from the antechamber tutorial. When I use bcc charge
>> model, the atomic types as well as the charges are recognized wrongly (see
>> am1bcc_berenil.mol2 attached): among other inconsistencies, one of the ring
>> is treated as a benzene ring, which is correct and another is not.
> This is certainly an interesting and unusual molecule. It is far enough
> away (I think) from molecules that were used in the development of GAFF
> that some weird behavior is not unexpected.
>
> It seems arguable to me that the two "benzene" rings are in fact in
> different environments. But it's easy enough to fix the atom types by
> hand, if you are convinced you know what they should be. One suggestion
> would be to minimize the structure in DFT or MP2, and look at the
> resulting bond lengths. That should provide a clue about what the best
> atom typing should be.
>
> If you are concerned about charges, be sure to check the resp charges
> for this molecule: resp charges don't depend on any atom typing, and are
> probably more reliable for rather unusual chemical species like this
> one.
>
> ...hope this helps....dac
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
Sergey Samsonov, PhD
Project Leader at Laboratory of Molecular Modeling,
Department of Theoretical Chemistry,
Faculty of Chemistry,
University of Gdansk
ul. Wita Stwosza 63
80-308 Gdańsk, Poland
phone: +48 58 523 5166
prot-gag.org
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Aug 17 2018 - 00:30:02 PDT
