we have tested tons of combinations.
In GPUS, with maybe 10 replicas or so, we can go down to nstlim=5 or 10
and we do not see any issues with slowing speed.
adrian
On 1/23/17 11:59 AM, Thakur, Abhishek wrote:
> Hi Adiran
>
> Thank you so much, I am using implicit water model with dt=0.004.
>
> I will refer to your papers.
>
> Has your group done testing with ntlism which does not slower the (ns/day)? If yes kindly let me know, so that I can start testing my system near to that figure.
>
> ________________________________
> From: Adrian Roitberg <roitberg.ufl.edu>
> Sent: Monday, January 23, 2017 5:50:29 AM
> To: AMBER Mailing List
> Subject: Re: [AMBER] REMD_suggestion
>
> Hi
>
> The answer depends a LOT on your setup.
>
> Are you doing this in explicit or implicit solvent ?
>
> My group has a couple of papers out (listed in my web page) where we
> show that you should try to exchange as often as you can. Basically,
> nstlim should be as small as you can make it. There is a cost associated
> with exchanges, so you should benchmark for nstlim=5, 10,100, for your
> system.
>
> You want to use the SMALLEST value of nstlim that does not make your
> simulation (ns/day) slower.
>
>
> Adrian
>
>
>
> On 1/23/17 11:46 AM, Thakur, Abhishek wrote:
>> Hi everyone,
>>
>> I am doing REMD simulation.
>>
>> My protein is a dimer and I need a suggestion at what time scale I should allow the exchange?
>>
>> I am trying to figure it out but bit confused what can be a goot time step.
>>
>>
>>
>> With regards,
>>
>> AT
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.ambermd.org_mailman_listinfo_amber&d=DwICAg&c=y2w-uYmhgFWijp_IQN0DhA&r=0Hah93XWYYBgmROOVcaccg&m=uJw0bDf1BI75KkbFmV7S0JXLmg4xwpQP1JHVPDBWU2g&s=gKdLfCSHmlwYidezxFatsGeKVYXnjF14IDyPkgzHgaI&e=
> --
> Dr. Adrian E. Roitberg
> University of Florida Research Foundation Professor.
> Department of Chemistry
> University of Florida
> roitberg.ufl.edu
> 352-392-6972
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.ambermd.org_mailman_listinfo_amber&d=DwICAg&c=y2w-uYmhgFWijp_IQN0DhA&r=0Hah93XWYYBgmROOVcaccg&m=uJw0bDf1BI75KkbFmV7S0JXLmg4xwpQP1JHVPDBWU2g&s=gKdLfCSHmlwYidezxFatsGeKVYXnjF14IDyPkgzHgaI&e=
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
--
Dr. Adrian E. Roitberg
University of Florida Research Foundation Professor.
Department of Chemistry
University of Florida
roitberg.ufl.edu
352-392-6972
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Jan 23 2017 - 09:30:03 PST
