To add to this. The GPU code does support GBSA calculations but the GBSA part of the GB calculation is run on the CPU so it is not as fast as running without the GBSA term.
All the best
Ross
> On Aug 5, 2016, at 9:31 AM, Carlos Simmerling <carlos.simmerling.gmail.com> wrote:
>
> Our gb model (igb=8) was trained using the Amber resp charge model. I would
> expect to have to do careful testing before using with a different charge
> set. My guess would be that it would not work well without some adjustment.
>
> I'm not positive but I don't think the CUDA code supports sasa
> calculations. Try it and see.
> Carlos Simmerling
>
> On Aug 5, 2016 12:06 PM, "Hai Nguyen" <nhai.qn.gmail.com> wrote:
>
>> On Fri, Aug 5, 2016 at 11:54 AM, Chitrak Gupta <chgupta.mix.wvu.edu>
>> wrote:
>>
>>> Hi,
>>>
>>> I am trying to do an implicit solvent simulation using pmemd.cuda (I am
>>> using amber16).
>>>
>>> One question is, do I need to set a specific surface tension?
>>
>>
>> amber turn off non-polar calculation for implicit solvent simulation as
>> default. So you don't need to specify surface tension.
>>
>>
>>> Some MD
>>> forums suggest that, but I did not find any reference to this in the
>> sample
>>> scripts given in the amber16 benchmarking page.
>>>
>>
>> I would suggest to use gbsa=0 (default). Please check previous discussion
>> in "[AMBER] gbsa parameter" topic.
>>
>>
>>>
>>> Another question is, is the GB solvent model related to the protein
>>> parameters? I am asking because I am trying to use CHARMM parameters in
>> my
>>> simulation. I have done this in the past with explicit solvent
>> simulations,
>>> but I am not sure if the implicit solvent has to be "consistent" with the
>>> force field parameters.
>>>
>>
>> You need to set radii 'bondi', 'mbondi2', 'mbondi3',... to define
>> solute/solvent boundary.
>> I am not familiar with CHARMM to AMBER topology conversion, so please wait
>> for others' feedback.
>>
>> Hai
>> _______________________________________________
>> 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
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Aug 05 2016 - 10:00:02 PDT
