On Wed, Aug 3, 2016 at 11:54 AM, Thomas Evangelidis <tevang3.gmail.com>
wrote:
> Greetings,
>
> The 'gbsa' parameter is set to 0 by default and when I set it to 1 or 2,
> the GB simulations as slowed down.
Indeed. Especially if you're running AMBER on GPU,gbsa will slow your
simulation down at least one order of magnitude.
The non-polar model (gbsa=1) is not that accurate while you can not run MD
with gbsa=2 (as far as I can recall).
> Is the SASA calculation necessary,
>
I think it's necessary if it's accurate.
> namely have the igb=1, 2, 5, 7 or 8 been optimized using gbsa=0 or not?
>
igb=2, 5, 7, 8 were optimized independently from non-polar energy (GB
energies were fitted to PB solvation energies).
You might want to check our protein folding paper (up to ~50 microsecond
REMDs): http://pubs.acs.org/doi/abs/10.1021/ja5032776
We excluded gbsa in the favor of GPU speed.
>
> Another question. OpenMM calculates by default the SASA using the "ACE
> approximation". Should I turn this feature off to get consistent results
> when I run GB simulations with AMBER .prmtop files (provided I used option
> gbsa=0 in AMBER simulations)?
>
>
you can try to see if openmm can reproduce amber energies if turning off
ACE.
FYI (in case you don't know): You can use ParmEd and pysander for energy
evaluation when comparing to openmm energies.
https://github.com/swails/OpenMMEnergyComparisons/blob/master/Amber/compare.py
Per turning on/off ACE, it's better to do literature search about its
accuracy.
Hai
>
> thanks in advance
> Thomas
>
>
> --
>
> ======================================================================
>
> Thomas Evangelidis
>
> Research Specialist
> CEITEC - Central European Institute of Technology
> Masaryk University
> Kamenice 5/A35/1S081,
> 62500 Brno, Czech Republic
>
> email: tevang.pharm.uoa.gr
>
> tevang3.gmail.com
>
>
> website: https://sites.google.com/site/thomasevangelidishomepage/
> _______________________________________________
> 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 Wed Aug 03 2016 - 10:00:02 PDT
