> On Aug 19, 2016, at 7:35 AM, DmitryASuplatov <genesup.gmail.com> wrote:
>
> Dear Amber users,
>
> I understand that the larger is the cutoff for nonbonded interactions
> ("cut" parameter) the better,
This is actually not as true as you would think.
For electrostatic interactions, the cutoff only changes the overall contributions of the direct and reciprocal space calculations -- the sum of the two won't change.
For lennard jones, we use a hard cutoff, but that falls off as 1/r^6, which is a lot faster than 1/r. Moreover, we use a fairly accurate analytical correction to get the truncated energy contribution.
For PME calculations, I think most people tend to use between 8-10 angstroms, but very rarely more. For GB calculations, we tend to use an infinite cutoff.
HTH,
Jason
> however, larger values dramatically slows
> down the calculations. So its really the question of what resources you
> have. Since the computer power gets more affordable every year I assume
> that the "normal" cutoff, i.e. what users typically set to make their
> simulations supposedly more realistic, also increases. So my aim is to
> understand the current "normal" average.
>
> I myself use Tesla K40s accelerators. I set the cut=12 and then make
> water box with at least 15A gap between the protein and the cell edges
> (to avoid situations when the protein interacts with itself from another
> cell). With these settings four GPU units provide an average speed of
> approximately 8-40 ns/day for systems of 190 000 - 70 000 atoms,
> respectively (in NVT). Using more units does not improve the
> performance, so this is the maximum I can get. This speed is comfortable
> for my tasks and so I stick with these settings.
>
> How do you set you cutoff and box gap? Would you use a larger cut if you
> could?
>
> Thanks for your time
> Dmitry
>
>> On 08/18/2016 10:00 PM, amber-request.ambermd.org wrote:
>> Send AMBER mailing list submissions to
>> amber.ambermd.org
>>
>> To subscribe or unsubscribe via the World Wide Web, visit
>> http://lists.ambermd.org/mailman/listinfo/amber
>> or, via email, send a message with subject or body 'help' to
>> amber-request.ambermd.org
>>
>> You can reach the person managing the list at
>> amber-owner.ambermd.org
>>
>> When replying, please edit your Subject line so it is more specific
>> than "Re: Contents of AMBER digest..."
>>
>>
>> AMBER Mailing List Digest
>>
>> Today's Topics:
>>
>> 1. Problem when visualizing bonds in VMD (Xiaoliu Zhang)
>> 2. Re: Problem when visualizing bonds in VMD (Jason Swails)
>> 3. Re: Is there any CG martini parameter for K+? (Pengfei Li)
>> 4. QM region has an odd number of electrons (Corum, Katharine W)
>> 5. Re: QM region has an odd number of electrons
>> (Dr. Andreas W. Goetz)
>> 6. RMSd (Wong Li Zhe)
>> 7. Re: RMSd (Hai Nguyen)
>> 8. Re: RMSd (Wong Li Zhe)
>> 9. Re: RMSd (Wong Li Zhe)
>> 10. Re: MD simulation with non natural amino acid, parameterized
>> using the RED server (Cecilia Lindgren)
>> 11. Re: RMSd (Bill Ross)
>> 12. Re: MD simulation with non natural amino acid, parameterized
>> using the RED server (Bill Ross)
>> 13. Re: solid surface simulation in amber (Neha Gandhi)
>> 14. Re: solid surface simulation in amber (Bill Ross)
>> 15. Re: MD simulation with non natural amino acid, parameterized
>> using the RED server (Cecilia Lindgren)
>> 16. Re: solid surface simulation in amber (Neha Gandhi)
>> 17. Re: MD simulation with non natural amino acid, parameterized
>> using the RED server (Bill Ross)
>> 18. LEaP and PARMSETs (Aronica, Pietro)
>> 19. Re: solid surface simulation in amber (Bill Ross)
>> 20. Re: MD simulation with non natural amino acid, parameterized
>> using the RED server (Cecilia Lindgren)
>> 21. Re: MD simulation with non natural amino acid, parameterized
>> using the RED server (Bill Ross)
>> 22. Re: MD simulation with non natural amino acid, parameterized
>> using the RED server (Hannes Loeffler)
>> 23. a bug in leaprc.gaff2 file (Ye Mei)
>> 24. Re: MD simulation with non natural amino acid, parameterized
>> using the RED server (Bill Ross)
>> 25. Re: MD simulation with non natural amino acid, parameterized
>> using the RED server (Cecilia Lindgren)
>> 26. Re: [AMBER} NetCDF installation problem on mac OS X El
>> Capitan (Jason Swails)
>> 27. Re: solid surface simulation in amber (David A Case)
>> 28. Re: a bug in leaprc.gaff2 file (David A Case)
>> 29. Re: a bug in leaprc.gaff2 file (David A Case)
>> 30. Problem in charge generation through antechamber (ATUL KUMAR)
>>
>>
>> ----------------------------------------------------------------------
>>
>> Message: 1
>> Date: Wed, 17 Aug 2016 19:40:22 +0000
>> From: Xiaoliu Zhang <xzhan91.lsu.edu>
>> Subject: [AMBER] Problem when visualizing bonds in VMD
>> To: "amber.ambermd.org" <amber.ambermd.org>
>> Message-ID:
>> <BY1PR0601MB133393B49EDBEFFBD1EBE94CA7140.BY1PR0601MB1333.namprd06.prod.outlook.com>
>>
>> Content-Type: text/plain; charset="iso-8859-1"
>>
>> Dear Amber users,
>>
>>
>> I'm using amber with Amoeba force field following the Amber 16 manual, chapter 17.
>>
>>
>> After I got prmtop and inpcrd file by using tinker_to_amber command, I tried VMD to visualize my molecule. My Ambertool is 16 version and VMD is 1.9.1.
>>
>>
>> First i got an error:
>>
>> AMBER 7 parm read error at flag section TITILE.
>>
>> expected format %FORMAT(20a4) but got %FORMAT(a)
>>
>>
>> Then, I checked my topology file and found that it begins with this:
>>
>>
>> %VERSION VERSION_STAMP = V0001.000 DATE = 08/17/16 08:18:25
>> %FLAG TITLE
>> %FORMAT(a)
>> 7box_5 ameoba FF
>> %FLAG POINTERS
>> %FORMAT(10I8)
>>
>> I changed the %FORMAT(a) to %FORMAT(20a4) by hand and ran again.
>>
>>
>> This time there's no error in terminal. However, no matter I choose VDW or CPK, there are only atoms showing on the screen, not any bond information at all.
>>
>>
>> Do you know how to fix this problem?
>>
>>
>> Thanks in advance.
>>
>>
>> Xiaoliu
>>
>>
>>
>> ------------------------------
>>
>> Message: 2
>> Date: Wed, 17 Aug 2016 16:17:27 -0400
>> From: Jason Swails <jason.swails.gmail.com>
>> Subject: Re: [AMBER] Problem when visualizing bonds in VMD
>> To: AMBER Mailing List <amber.ambermd.org>
>> Message-ID:
>> <CAEk9e3r8uYSuqwK3=hpfdsrD457wqbJcZ0cBGy_mq=H9-bEQxA.mail.gmail.com>
>> Content-Type: text/plain; charset=UTF-8
>>
>>> On Wed, Aug 17, 2016 at 3:40 PM, Xiaoliu Zhang <xzhan91.lsu.edu> wrote:
>>>
>>> Dear Amber users,
>>>
>>>
>>> I'm using amber with Amoeba force field following the Amber 16 manual,
>>> chapter 17.
>>>
>>>
>>> After I got prmtop and inpcrd file by using tinker_to_amber command, I
>>> tried VMD to visualize my molecule. My Ambertool is 16 version and VMD is
>>> 1.9.1.
>>>
>>>
>>> First i got an error:
>>>
>>> AMBER 7 parm read error at flag section TITILE.
>>>
>>> expected format %FORMAT(20a4) but got %FORMAT(a)
>>>
>>>
>>> Then, I checked my topology file and found that it begins with this:
>>>
>>>
>>> %VERSION VERSION_STAMP = V0001.000 DATE = 08/17/16 08:18:25
>>> %FLAG TITLE
>>> %FORMAT(a)
>>> 7box_5 ameoba FF
>>> %FLAG POINTERS
>>> %FORMAT(10I8)
>>>
>>> I changed the %FORMAT(a) to %FORMAT(20a4) by hand and ran again.
>>>
>>>
>>> This time there's no error in terminal. However, no matter I choose VDW
>>> or CPK, there are only atoms showing on the screen, not any bond
>>> information at all.
>>>
>>>
>>> Do you know how to fix this problem?
>> ?You can't use AMOEBA topology files with VMD easily.
>>
>> Try using a PDB as the topology instead.
>>
>> HTH,
>> Jason
>
>
> _______________________________________________
> 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 19 2016 - 06:00:04 PDT
