[gmx-users] configuration does not change in minimization trajectory

[Justin A. Lemkul]

heiko252 at web.de wrote:
>> -----Ursprüngliche Nachricht-----
>> Von: "Justin A. Lemkul" <jalemkul at vt.edu>
>> Gesendet: 01.05.09 03:10:01
>> An: Discussion list for GROMACS users <gmx-users at gromacs.org>
>> Betreff: Re: [gmx-users] configuration does not change in minimization trajectory
> 
> 
>>
>> heiko252 at web.de wrote:
>>> Hello all,
>>>
>>> I am doing a cg minimization with Gromacs-4.0.3.
>>> It is compiled on a Linux system with gcc 3.4.6
>>>
>>> All frames in the trajectory are identical to the starting configuration,
>>> except for the final one, which is clearly different.
>>>
>> How are you assessing this?  External viewing programs don't always show the 
>> necessary level of detail.  If you use trjconv -dump to drop out a few select 
>> frames, and diff the resulting .gro files, are they indeed identical, or are 
>> there small differences?
>>
> All configurations but the last one are strictly identical to the first.
> (Determined as suggested, using diff on .gro files dumped from the trajectory)
> 

I see something similar with my own CG trajectories.  I do not routinely inspect 
them; I typically rely on the energy curve and final output structure to 
determine the adequacy of the procedure.

What I find in my own case is that the third decimal differs by no more than 1 
unit between the reference structure and the constituent frames of the .trr 
file.  It would seem to me that if you are seeing no difference along the 
different frames, this is probably just a machine precision issue.

Differences are more obvious, however, when using steepest descents, and the 
trajectory, when viewed in VMD, shows continual change in the structure.

>> How different is the final structure?  Is there a drastic change?  Or is it just 
>> that the change relative to the input structure is more obvious?
>>
> The rmsd between the first and last structures is about 0.1 nm,
> using -nomw with g_rms. So any changes to the structure are local, but obvious in VMD.
> 

I would think that 0.1 nm is somewhat large for a simple minimization procedure, 
although this is probably dependent upon system size and which components you 
are analyzing.

-Justin

>>> I get the expected number of frames in the trajectory (.trr),
>>> and the energy saved in the .edr is decreasing.
>>> But apparently the changes in the structure do not reach the trajectory file.
>>>
>> If the energy is changing, then so too should the structure.
>>
> I agree, internally the structure must be changing, only the output does not show it.
> 
>> -Justin
>>
>>> Any suggestions?
>>>
>>> Heiko
>>>
>> Justin A. Lemkul
>> Ph.D. Candidate
>> ICTAS Doctoral Scholar
>> Department of Biochemistry
>> Virginia Tech
>> Blacksburg, VA
>> jalemkul[at]vt.edu | (540) 231-9080
>> http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
> 
> Thanks for your input,
> 
> Heiko
> 
> ____________________________________________________________________
> Psssst! Schon vom neuen WEB.DE MultiMessenger gehört? 
> Der kann`s mit allen: http://www.produkte.web.de/messenger/?did=3123
> 
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at http://www.gromacs.org/search before posting!
> Please don't post (un)subscribe requests to the list. Use the 
> www interface or send it to gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
> 

-- 
========================================

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================