Hi,
I think your attachments were probably too large for the mail server.
Send them to me off-list if you would like. Again, without knowing
anything about your system I can only give general advice - maybe try
a longer minimization in step 1 and/or change to a different algorithm
(try XMIN, imin=3). Pay attention to your energy gradient, you
probably want to get it in the E-2 range or smaller.
-Dan
On Sun, Jul 21, 2013 at 5:27 PM, John Travers <jtravers70.yahoo.com> wrote:
> Dear Dr Roe
>
> Thank you so much for the explanation and instructions. I followed your procedure, and unfortunately got stuck at step 2. The MD stopped at the 5th step (segmentation fault) and the vlimit was becoming larger and larger. Attached are the parameter files and the related inputs. Do you have other suggestions? Thank you again.
>
> Best wishes
>
> JT
>
>
> ________________________________
> From: Daniel Roe <daniel.r.roe.gmail.com>
> To: AMBER Mailing List <amber.ambermd.org>
> Sent: Sunday, July 21, 2013 2:57 PM
> Subject: Re: [AMBER] problem on heating with restraints on the solute
>
>
> Hi,
>
> On Sun, Jul 21, 2013 at 12:30 PM, John Travers <jtravers70.yahoo.com> wrote:
>> Yes, I am talking about tutorial B1, step 5.1.5 but with the molecule of
>> mine.
>
> The tutorials are meant as guidelines only, not as templates for how
> to do minimization, equilibration, etc. The fact that your kinetic
> energy goes all NaN within a few steps indicates that your previous
> minimization/equilibration was inadequate. Without knowing more
> details of your system I can only make general recommendations, but
> for an explicitly solvated system I usually follow a 9 step initial
> minimization/equilibration scheme:
>
> 1) 1K step Steepest Descent Minimization with strong (5
> kcal/mol*Ang^2) restraints on heavy atoms, no shake.
> 2) NTV MD with strong restraints on heavy atoms and shake, dt=.001, 15 ps.
> 3) 1K step Steepest Descent Minimization with relaxed (2
> kcal/mol*Ang^2) restraints on heavy atoms, no shake.
> 4) 1K step Steepest Descent Minimization with minimal (0.1
> kcal/mol*Ang^2) restraints on heavy atoms, no shake
> 5) 1K step Steepest Descent Minimization with no restraints, no shake.
> 6) NTP MD with shake and low (1.0 kcal/mol*Ang^2) restraints on heavy
> atoms, 5 ps dt=.001.
> 7) NTP MD with shake and minimal (0.5 kcal/mol*Ang^2) restraints on
> heavy atoms, 5 ps dt=.001.
> 8) NTP MD with shake and minimal restraints on backbone atoms, dt=0.001, 10 ps
> 9) NTP MD with shake and no restraints, dt=0.002, 10 ps.
>
> For every MD step temp0 is always set to the temperature of choice,
> and I use a Berendsen thermostat/barostat. Of course, all restraints
> are applied to the solute molecule only. This procedure is usually
> good enough to get you on your way - I generally run out step 9 until
> the density (among other things) seems well-behaved before even
> considering a production run. For steps 1-4 I use as reference
> structure the initial coordinates for that step, and for steps 6-8 I
> use the final structure from step 5. Others probably have their own
> minimization schemes but this is one that works for me on most
> systems.
>
> Hope this helps,
>
> -Dan
>
>>
>> Best
>>
>> JT
>>
>> ________________________________
>> From: Daniel Roe <daniel.r.roe.gmail.com>
>> To: John Travers <jtravers70.yahoo.com>; AMBER Mailing List
>> <amber.ambermd.org>
>> Sent: Saturday, July 20, 2013 6:17 PM
>> Subject: Re: [AMBER] problem on heating with restraints on the solute
>>
>> Hi,
>>
>> Are you talking about tutorial B1, step 5.1.5? Are you using the files
>> provided from the website ( polyAT_wat_md1.in, polyAT_wat.prmtop,
>> polyAT_wat_min2.rst)?
>>
>> -Dan
>>
>> On Sat, Jul 20, 2013 at 2:21 PM, John Travers <jtravers70.yahoo.com> wrote:
>>> Dear Amber experts
>>>
>>> I am a new user of AMBER. I am trying to run a MD simulation following the
>>> tutorial on simulating a DNA polyA-polyT Decamer. I got a problem at the
>>> heating stage. I used the SHAKE algorithm (the same as in the tutorial), but
>>> the MD only ran for 7 steps (stepsize 1fs) and stopped with the error
>>> message of segmentation fault. At the step 7, the kinetic energy term
>>> becomes NaN. On the other hand, if I don't use the SHAKE algorithm (ntc=1,
>>> ntf=1, time step=0.2 fs) the MD ran normally. Any hints on this?
>>>
>>> I also check the 6 step MD results using VMD, it seemed only a small part
>>> of the whole water was moving. I followed strictly the tutorial. The partial
>>> and full minimization went smoothly. I compared the structures before and
>>> after minimization, they looked just fine.
>>>
>>> Thanks!
>>>
>>> Best wishes
>>>
>>> JT
>>> _______________________________________________
>>> AMBER mailing list
>>> AMBER.ambermd.org
>>> http://lists.ambermd.org/mailman/listinfo/amber
>>
>>
>>
>> --
>> -------------------------
>> Daniel R. Roe, PhD
>> Department of Medicinal Chemistry
>> University of Utah
>> 30 South 2000 East, Room 201
>> Salt Lake City, UT 84112-5820
>> http://home.chpc.utah.edu/~cheatham/
>> (801) 587-9652
>> (801) 585-9119 (Fax)
>>
>
>
>
> --
> -------------------------
> Daniel R. Roe, PhD
> Department of Medicinal Chemistry
> University of Utah
> 30 South 2000 East, Room 201
> Salt Lake City, UT 84112-5820
> http://home.chpc.utah.edu/~cheatham/
> (801) 587-9652
> (801) 585-9119 (Fax)
>
> _______________________________________________
> 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
--
-------------------------
Daniel R. Roe, PhD
Department of Medicinal Chemistry
University of Utah
30 South 2000 East, Room 201
Salt Lake City, UT 84112-5820
http://home.chpc.utah.edu/~cheatham/
(801) 587-9652
(801) 585-9119 (Fax)
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sun Jul 21 2013 - 18:00:02 PDT
