From: Thomas Evangelidis (tevang3_at_gmail.com)
Date: Wed Apr 13 2011 - 12:06:57 CDT
There's a non-trivial issue when saving water and other rapidly moving
hetatms using their atom indices: as the simulation progresses these atoms
will move far away from your macromolecule thus leaving it unsolvated. The
right way is setting a constant number of waters, lets say N, and NAMD
should keen only the N closest waters in each frame. The same applies to
other hetatms like ions. In order to do that NAMD requires a functions that
measures the number of waters within a given volume, and must expand or
contract that volume until it finds the one that encapsulates N waters. VMD
has such a function implement by "within" keyword of "atomselect" command,
but I'm not sure about NAMD. Nevertheless, I had started developing a plugin
fron VMD in the past to truncate trajectories in that fashion. I hope to
finalize a stable version for the current VMD version sometime in 2011.
Here's a description of what it could do:
https://sites.google.com/site/thomasevangelidishomepage/truncate-trajectory-plugin-for-vmd
2011/4/13 Giacomo Fiorin <giacomo.fiorin_at_gmail.com>
> Hi Ajasja, you can compress the colvars.traj files quite a lot if you run
> gzip or bzip2 on them. This way you can still analyze them with scripts (by
> processing them on the fly with gzip -dc), rather than write specific
> programs to even open them. To save extra space, have you considered doing
> what you propose for the DCD trajectories, i.e. keeping only a subset of the
> variables at high frequency, and instead the whole set at much lower
> frequency?
>
> Saving the DCD files for a subset of the atoms can be useful in a few
> specific cases, but on a general basis I would advise against it. Not
> knowing at all what's happening to the lipid membrane and the solvent seems
> a bit risky. At least the way you're doing now you can look at the whole
> system and see that it looks OK before you can neglect the uninteresting
> part.
>
> Giacomo
>
> ---- ----
> Giacomo Fiorin - Postdoctoral Researcher
> ICMS - Institute for Computational Molecular Science - Temple University
> 1900 N 12 th Street, Philadelphia, PA 19122
> giacomo.fiorin_at_temple.edu
> ---- ----
>
>
>
>
> 2011/4/13 Ajasja Ljubetič <ajasja.ljubetic_at_gmail.com>
>
>> Dear NAMD developers,
>>
>> I have two feature request which are related, perhaps even mutually
>> exclusive:
>>
>> - Would it be possible to add binary output for colvars trajectories?
>> I would like to write several colvars at high frequency and the text
>> files are getting into Gigabytes.
>>
>>
>>
>> - Would it be possible to implement writing only part of the system
>> into the binary dcd trajectory.
>> I have a medium membrane system, but I'm interested only in a small
>> specific part. It would be ideal if NAMD could read a file of atom indices
>> and write only those atoms to the dcd. (And I know I can run my simulation
>> in parts and then load the trajectory in VMD and save only the interesting
>> part and delete the oiginal dcd - this is what I'm doing now.)
>>
>> Best regards,
>> Ajasja Ljubetič,
>> Young reasercher,
>> Laboratory of biophysics,
>> Institute Jožef Štefan,
>> Ljubljana, Slovenia
>>
>>
>
>
--
======================================================================
Thomas Evangelidis
PhD student
Biomedical Research Foundation, Academy of Athens
4 Soranou Ephessiou , 115 27 Athens, Greece
email: tevang_at_bioacademy.gr
tevang3_at_gmail.com
website: https://sites.google.com/site/thomasevangelidishomepage/
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:56:59 CST