Hi Satyajit,
When you calculate MMPB/GBSA results for metalloproteins, if the metal center does not change during the ligand binding, then it is fine for you to use a bonded model parameterized by MCPB.py. If the metal center does change during the ligand binding, such as ligand will bind to the metal ion during the binding process, you should not use a bonded model because it can not from/break metal-ligand bonds, in which case a nonbonded model would be more suitable. (You can check this tutorial about the difference of bonded and nonbonded models: https://ambermd.org/tutorials/advanced/tutorial20/index.htm. ) However, as Prof. David Case pointed out, the PB/GB results for divalent ions depend very strongly on the dielectric radii that are used, so for the later case I am not sure how meaningful the obtained results are. This issue may be insignificant for the former case when you use the bonded model as in the bonded model the metal ion usually has a charge less than its formal charge.
Hope it helps,
Pengfei
> On Nov 5, 2020, at 8:37 AM, David A Case <david.case.rutgers.edu> wrote:
>
> On Thu, Nov 05, 2020, SATYAJIT KHATUA wrote:
>>
>> Is it possible to run the minimization using sander.MPI instead of the
>> sander module? As in the tutorial, they only use sander for all the steps
>> before going to the production run.
>
> What happened when you tried this? It's possible that there is some bug
> in sander.MPI that affects minimization, since the MPI code for
> minimization is very different than that for MD. But without any
> details, it is hard to be of much help.
>
> ....dac
>
> [Note that the tutorials show one way to carry out certain tasks. Don't
> be afraid to experiment! In particular, tutorials are usually designed
> to be done quickly; they may not represent best practices for "real"
> simulations.]
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Nov 20 2020 - 08:30:03 PST
