[gmx-users] restartng run..........

[Justin A. Lemkul]

On 1/10/10 12:08 AM, pavan payghan wrote:
> HI
>         I want to run my system for almost 50 ns but as diskspace
> becomes almost full to run system for this much time
>         so which of the option u suggest
>         1. Giving 10ns with each run by .mdp option .
>         2.  giving full run of 50ns and then creating at each time new
> tpr file to continue run for next step while storing .xtc
>             at each step and concurrent removal of .trr file as it
> occupies a lots of diskspace.
>
>       While restarting run by using tpbconv to create new .tpr file for
> new run may i know  following of which commond is
>       correct
>
>            1.  tpbconv -s md1_out.tpr -o md2_out.tpr -e md1_out.edr
>             2  tpbconv -s md1_out.tpr -o md2_out.tpr -e md1_out.edr
> -extend 8000.

I don't know what #1 would actually be doing; seems like nothing.

>             tell me what -extends signifies i.e.point from where you
> want to continue or extend the given number of steps.

 From tpbconv -h:

"-extend      real   0       Extend runtime by this amount (ps)"

>        im asking this because when i run it using previous (1) commend i
> got run in which the middle portion was
>        containing no information about run.
>        that is system run well till 7 ns then i killed run to continue
> it for seond time this time it run well but at the end
>        i found that 7ns - 13ns portin was  not there, why it is so?
>       ( may it be the reason because some incomplete frames are present .)
>      also
>       tell me which is the best way to check whether the system is
> running well or not .?

I think you need to read about the checkpointing feature, it simplifies all of 
this immensely:

http://www.gromacs.org/Documentation/How-tos/Extending_Simulations

Very simple.

>       regarding g_rms
>       While selecting groups for RMSD which is correct
>       1. select whole system with solvent
>        2. select only molecule of interest .

The selection depends on which step you're talking about (least squares fitting 
or actual calculation).  See below.  I doubt you ever want to consider solvent 
for anything, though (assuming you're simulating some "normal" protein system).

>        also while choosing groups is it correct to choose both time i.e.
> for least square fitting and RMSD same groups.
>

It certainly can be, depending on which groups you want to use for the fitting. 
  Best to do some background reading on what these algorithms are doing before 
just haphazardly making choices :)

-Justin

>       Thanking you in Advance.
>
>        Mr. Pavan Payghan
>        INDIAN INSTITUTE OF CHEMICAL BIOLOGY
>        KOLKATA
>

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

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

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