[gmx-users] the -extend function of the tpbconv command

[Mark Abraham]

On 4/04/2012 6:24 PM, Acoot Brett wrote:
> Hi Justin and All,
> The on-line document on "Extending Simulations" writes,
> "You might want to use the -append option of mdrun 
> <http://www.gromacs.org/Documentation/Gromacs_Utilities/mdrun> to 
> append the new output to the old files. Note that this will only work 
> when the old output files have not been modified by the user. 
> Appending is the default behavior as of version 4.5."
> We can see "Appending is the default behavior as of version 4.5", and 
> what I use is version 4.5. Thus for mdrun, even without mention 
> "-append " appending should be automatically done.
> In addition, after the old output files were created, I only use the 
> "trjconv" to process them, and I did not modify them.

Any change is enough to prevent appending. mdrun computed a checksum 
when it wrote them, saved the checksums in the checkpoint file, and 
demands that newly-computed ones match when attempting to append later. 
Use only the original trajectory files when attempting to append.

Mark

> Thus the reason why the extended part output files were not 
> automatically appended to my original old outputfiles according to 
> what I have done (tpbconv -s previous.tpr -extend timetoextendby -o 
> next.tpr       mdrun -s next.tpr -cpi previous.cpt)is still unclear.
> Will you please give an explaination?
> Cheers,
> Acoot
>
> *From:* Justin A. Lemkul <jalemkul at vt.edu>
> *To:* Acoot Brett <acootbrett at yahoo.com>; Discussion list for GROMACS 
> users <gmx-users at gromacs.org>
> *Sent:* Wednesday, 4 April 2012 10:36 AM
> *Subject:* Re: [gmx-users] the -extend function of the tpbconv command
>
>
>
> Acoot Brett wrote:
> > Dear All,
> >  I have completed a 1ns production MD. Then I intended to extend it 
> for another 1 ns.
> >  I use command
> >  "tpbconv -s previous.tpr -extend timetoextendby -o next.tpr
> > mdrun -s next.tpr -cpi previous.cpt"
> >  I think the results will be automatically attended to previous 
> md_0_1.tpr and md_0_1_noPBC.xtc.
> >
>
> Only if you then run mdrun -append and the files are appended.  
> tpbconv doesn't do any of this for you, it just creates a new run 
> input file capable of running a longer simulation.  The fact that you 
> have a file called "md_0_1_noPBC.xtc" seems like you've manipulated it 
> in some way, so you shouldn't (and likely can't) append to it.  You 
> should append to your original files.
>
> > Then I use "g_rms -s md_0_1.tpr -f md_0_1_noPBC.xtc -o rmsd.xvg -tu 
> ns" to get the rmsd.xvg file. Then I use "xmgrace -nxy potential.xvg" 
> to display it. However I find for the  curve the x-axis is from 0 ns 
> to 1 ns, not from 0 ns to 2 ns.
> >
>
> If you haven't run a new simulation, you'll only have the data that 
> already exists.
>
> > The definition of "-extend" of the command "tpbconv" is "Extend 
> runtime by this amount (ps)". Thus the commands I used are correct.
> >
>
> You haven't shown us your actual tpbconv command, so there's no way to 
> verify that.  In any case, tpbconv doesn't actually extend the 
> simulation, mdrun does.
>
> > Will you please tell me how to process for the extension MD so that 
> the curve of the rmsd.xvg (and all the other xvg files) will have 
> x-axis from the 0 to the 2 ns (2 ns is the total MD time, including 
> the 1 ns extended part)?
> >
>
> There are two steps here:
>
> http://www.gromacs.org/Documentation/How-tos/Extending_Simulations#Version_4
>
> You need to do both.
>
> -Justin
>
> -- ========================================
>
> 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
>
> ========================================
>
>
>
>
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20120404/67e226b5/attachment.html>