<div>I am doing a manual replica exchange(generalized hamiltonian rem) after every mdrun. If the replica exchange is successful, then I "exchange checkpoint files". For example, consider the following:</div>
<div> </div>
<div>Simulation parameters:....................B1.........B2</div>
<div>Replicas(coordinates and velocities):.R1.........R2</div>
<div> </div>
<div>
<div>0. Tpbconv to extend simulation time ( using -s, -o and -nsteps ONLY)</div>
<div>1. Mdrun run - 500 steps = 1ps</div>
<div>2. Attempt exchange - NOT SUCCESSFUL</div>
<div>3. Exchange implementation: SKIP </div>
<div>4.Continue to next step</div></div>
<div> </div>
<div>.....</div>
<div>0. Tpbconv to extend simulation time ( using -s, -o and -nsteps ONLY)</div>
<div>1. Mdrun run - 500 steps = 1ps</div>
<div>2. Attempt exchange - If successful, exchange</div>
<div>3. Exchange Implemented by - exchanging checkpointing files</div>
<div>4. Continue to next step</div>
<div> </div>
<div>0. Tpbconv to extend simulation time ( using -s, -o and -nsteps ONLY)</div>
<div>1. Mdrun with exchanged .cpt files -NOW this is where the problem shows.. </div>
<div> i) The log, xtc files are not appended when beginning after a step with a successful exchange attempt:<em>According to Mark's previous mail, this could be a result of mismatch in ensembles. Which means that the .cpt is ignored - implying that the mdrun in B1 DOES NOT start from R2.</em></div>
<div> </div>
<div>Therefore, to make R1 run in B2 and R2 run in B1, do I need to supply .cpt to tpbconv instead of mdrun after a successful exchange step?</div>
<div> </div>
<div>To summarize:</div>
<div>APPENDING HAS NOW BECOME A SECONDARY CONCERN, WHAT I AM INTERESTED IN IS A SUCCESSFUL MANUAL REPLICA EXCHANGE RUN AS POINTED OUT ABOVE.</div>
<div> </div>
<div>I hope my dilemma is clear now.</div>
<div> </div>
<div>Pooja</div>
<div> <br><br></div>
<div class="gmail_quote">On Tue, Feb 1, 2011 at 11:42 AM, Justin A. Lemkul <span dir="ltr"><<a href="mailto:jalemkul@vt.edu">jalemkul@vt.edu</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">
<div class="im"><br><br>Sai Pooja wrote:<br>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid"> From the website:<br> "If you change the integrator or ensemble, you should pass the checkpoint file to tpbconv only, not to mdrun, since the state might change and thus output files can not be appended."<br>
<br></blockquote><br></div>Are you changing the integrator, ensemble, and/or other settings? If not, this statement does not apply. For a simple -append after extending via tpbconv, this is not applicable.
<div class="im"><br><br>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">So now instead of supplying the checkpoint file to mdrun I supply it to tpbconv... does this assure that the simulations start from the coordinates/velocities specified by the .cpt file?<br>
<br></blockquote><br></div>For a simple extension, you do not pass the .cpt file to tpbconv.<br><br><a href="http://www.gromacs.org/Documentation/How-tos/Extending_Simulations#Version_4" target="_blank">http://www.gromacs.org/Documentation/How-tos/Extending_Simulations#Version_4</a>
<div class="im"><br><br>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid"> Thanks Mark. So if the simulation doesn't start from the checkpoint<br> file, from where are the initial coordinates velocities etc. taken<br>
from?... the trajectory files?<br></blockquote><br></div>Mark's previous message already answered this.<br><br>-Justin<br><br>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">
<div class="im"> Also, I could not find the environment variable... and I am not sure<br> how to use one.<br> Pooja<br> On Tue, Feb 1, 2011 at 3:03 AM, Mark Abraham<br></div>
<div>
<div></div>
<div class="h5"> <<a href="mailto:Mark.Abraham@anu.edu.au" target="_blank">Mark.Abraham@anu.edu.au</a> <mailto:<a href="mailto:Mark.Abraham@anu.edu.au" target="_blank">Mark.Abraham@anu.edu.au</a>>> wrote:<br>
<br> On 1/02/2011 7:50 AM, Sai Pooja wrote:<br>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid"> I think I have figured out the reason. It is because I am<br> carrying out replica exchange (manual) after every mdrun. If<br>
the exchange occurs, I exchange the checkpoint files, extend<br> the simulation by 500 steps and continue. The new simulation<br> starts from exchanged cptfile. It seems that whenever the<br> exchange occurs, the earlier log,traj files are not appended.<br>
They are instead overwritten. the obv solution is to save and<br> index these files with the relevant replicas everytime an<br> exchange occurs.<br></blockquote><br> This would have been good to know earlier. If replica-exchange<br>
leads to the ensemble of the .tpr not matching the ensemble of<br> the .cpt, then IIRC 4.5.3 mdrun will refuse to start from the<br> .cpt, which means the subsequent mdrun will start from the .tpr<br> only. Certainly a non-appending mdrun prints a warning (or<br>
error, I forget which) message to the log file, but perhaps the<br> use of -append (erroneously) doesn't do that. Please have a look<br> and see if that is the issue. There is an environment variable<br>
that can be set to tell mdrun that you (think you) know what you<br> are doing mismatching .tpr and .cpt.<br><br> Mark<br><br><br></div></div>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">
<div>
<div></div>
<div class="h5"> However, i have a more general question. Since mdrun still<br> runs with the exchanged checkpoint files and starts from the<br> point where the previous run ended, can I be assured that an<br>
exchange has been affected - since tpr files correspond to the<br> "replica-box" and cpi to the most recent exchanged replicas?<br> Pooja<br><br> On Mon, Jan 31, 2011 at 2:33 PM, Sai Pooja <<a href="mailto:saipooja@gmail.com" target="_blank">saipooja@gmail.com</a><br>
</div></div>
<div class="im"> <mailto:<a href="mailto:saipooja@gmail.com" target="_blank">saipooja@gmail.com</a>>> wrote:<br><br><br> I manually index checkpoint files after every mdrun. What<br> troubles me is the randomness with which -append fails/works.<br>
For eg, I have a simulation which runs from 3ns, 1ps in 1<br> mdrun. Now oddly enough, the logfile starts from 1184ps(in<br> the end, I do remember the one starting from 0 but that<br>
was overwritten it seems) and the rest is appended uptil<br> the 3000ps step. Why would append work from 1184ps to<br> 3000ps but not for the previous ones?Could it have<br> anything to do with the network/cluster? If that is the<br>
case is it safer to create a new file everytime and then<br> concatenate them after say every 100ps?<br> Pooja<br> On Sat, Jan 29, 2011 at 6:52 PM, Mark Abraham<br></div>
<<a href="mailto:Mark.Abraham@anu.edu.au" target="_blank">Mark.Abraham@anu.edu.au</a> <mailto:<a href="mailto:Mark.Abraham@anu.edu.au" target="_blank">Mark.Abraham@anu.edu.au</a>>>
<div>
<div></div>
<div class="h5"><br> wrote:<br><br> On 30/01/2011 10:39 AM, Sai Pooja wrote:<br>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid"> I would be happy to supply more information.. if<br> someone could please look into this.. otherwise I<br>
will have to switch to storing every file and then<br> just concatenating them which seems like a rather<br> roundabout way of doing it.<br></blockquote><br> As I suggested a few emails ago, are you sure that<br>
-cpi file exists? If your numerical suffixes are<br> indexing restarts, then unless you've done some manual<br> copying that you haven't told us about, it won't. Your<br>
filename scheme seems a bit contorted - like you're<br> trying to do the work that GROMACS 4.5.x will just do<br> for you if you let it.<br><br> Otherwise, you'll have to do some detective work with<br>
gmxcheck on the -cpi to see what might be the issue.<br><br> In your case, an initial<br><br> mdrun -deffnm rex_3<br><br> (perhaps save some copies while you're experimenting)<br>
and subsequently<br><br> tpbconv -extend <blah> -f rex_3 -o rex_3<br> mdrun -deffnm rex_3 -append<br><br> will work and be much simpler than whatever you're<br>
trying to do with filenames :-)<br><br> Mark<br><br><br></div></div>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">
<div>
<div></div>
<div class="h5"><br> On Fri, Jan 28, 2011 at 4:37 PM, Sai Pooja<br></div></div>
<div class="im"> <<a href="mailto:saipooja@gmail.com" target="_blank">saipooja@gmail.com</a> <mailto:<a href="mailto:saipooja@gmail.com" target="_blank">saipooja@gmail.com</a>>> wrote:<br><br> This is the command:<br>
nbs submit -command<br> "(/usr/local/gromacs/4.5.1/bin/mdrun_mpi -s<br> rex_3.tpr -e rex_3 -c after_rex_3 -cpi restart3<br> -cpo restart3 -ap<br>
pend -g rexlog3 -x rextraj3);" -nproc 1 -name<br> "GENHAM-DIHEDRAL-3" -mail start end<br> Pooja<br><br> On Fri, Jan 28, 2011 at 4:20 PM, Mark Abraham<br>
<<a href="mailto:Mark.Abraham@anu.edu.au" target="_blank">Mark.Abraham@anu.edu.au</a><br></div>
<div>
<div></div>
<div class="h5"> <mailto:<a href="mailto:Mark.Abraham@anu.edu.au" target="_blank">Mark.Abraham@anu.edu.au</a>>> wrote:<br><br> On 29/01/2011 3:56 AM, Sai Pooja wrote:<br><br>
Hi,<br> I am using tpbconv and mdrun to extend a<br> simulation. I use it with the append<br> option but the files still get<br>
overwritten or erased. Can someone help<br> me in this regard?<br> Pooja<br> Commands (in python)<br> cmd = '(%s/tpbconv -extend %f -s<br>
rex_%d.tpr -o rex_%d.tpr)'<br> %(GROMPATH,dtstep,i,i)<br> os.system(cmd)<br> time.sleep(1)<br> cmd = 'nbs submit -command "'<br>
cmd += '(%s/mdrun_mpi -noh<br> -noversion -s rex_%d.tpr -e rex_%d -c<br> after_rex_%d -cpi restart%d -cpo<br> restart%d -append -g rexlog%d -x<br>
rextraj%d >/dev/null); '<br> %(GROMPATH,i,i,i,i,i,i,i)<br> cmd += '" '<br> cmd += '-nproc 1 '<br>
cmd += '-name "GENHAM-DIHEDRAL-%d" '%(i)<br> cmd += '-mail start end '<br> cmd += '-elapsed_limit 16h >> rexid'<br>
os.system(cmd)<br><br><br> More useful for diagnostic and<br> record-preservation purposes is to construct<br> the cmd string and print it to stdout (or<br>
something).<br><br> At the moment it is far from clear that your<br> -cpi file exists for the new run.<br><br> Mark<br> -- gmx-users mailing list <a href="mailto:gmx-users@gromacs.org" target="_blank">gmx-users@gromacs.org</a><br>
</div></div> <mailto:<a href="mailto:gmx-users@gromacs.org" target="_blank">gmx-users@gromacs.org</a>>
<div class="im"><br> <a href="http://lists.gromacs.org/mailman/listinfo/gmx-users" target="_blank">http://lists.gromacs.org/mailman/listinfo/gmx-users</a><br> Please search the archive at<br>
<a href="http://www.gromacs.org/Support/Mailing_Lists/Search" target="_blank">http://www.gromacs.org/Support/Mailing_Lists/Search</a><br> before posting!<br> Please don't post (un)subscribe requests to<br>
the list. Use the www interface or send it to<br> <a href="mailto:gmx-users-request@gromacs.org" target="_blank">gmx-users-request@gromacs.org</a><br></div> <mailto:<a href="mailto:gmx-users-request@gromacs.org" target="_blank">gmx-users-request@gromacs.org</a>>.
<div class="im"><br> Can't post? Read<br> <a href="http://www.gromacs.org/Support/Mailing_Lists" target="_blank">http://www.gromacs.org/Support/Mailing_Lists</a><br><br><br>
<br><br> -- Quaerendo Invenietis-Seek and you shall discover.<br><br><br><br><br> -- Quaerendo Invenietis-Seek and you shall discover.<br></div></blockquote>
<div class="im"><br><br> -- <br> gmx-users mailing list <a href="mailto:gmx-users@gromacs.org" target="_blank">gmx-users@gromacs.org</a><br></div> <mailto:<a href="mailto:gmx-users@gromacs.org" target="_blank">gmx-users@gromacs.org</a>>
<div class="im"><br> <a href="http://lists.gromacs.org/mailman/listinfo/gmx-users" target="_blank">http://lists.gromacs.org/mailman/listinfo/gmx-users</a><br> Please search the archive at<br> <a href="http://www.gromacs.org/Support/Mailing_Lists/Search" target="_blank">http://www.gromacs.org/Support/Mailing_Lists/Search</a><br>
before posting!<br> Please don't post (un)subscribe requests to the list.<br> Use the<br> www interface or send it to<br> <a href="mailto:gmx-users-request@gromacs.org" target="_blank">gmx-users-request@gromacs.org</a><br>
</div> <mailto:<a href="mailto:gmx-users-request@gromacs.org" target="_blank">gmx-users-request@gromacs.org</a>>.
<div class="im"><br> Can't post? Read<br> <a href="http://www.gromacs.org/Support/Mailing_Lists" target="_blank">http://www.gromacs.org/Support/Mailing_Lists</a><br><br><br><br><br> -- Quaerendo Invenietis-Seek and you shall discover.<br>
<br><br><br><br> -- Quaerendo Invenietis-Seek and you shall discover.<br></div></blockquote>
<div class="im"><br><br> -- <br> gmx-users mailing list <a href="mailto:gmx-users@gromacs.org" target="_blank">gmx-users@gromacs.org</a><br></div> <mailto:<a href="mailto:gmx-users@gromacs.org" target="_blank">gmx-users@gromacs.org</a>>
<div class="im"><br> <a href="http://lists.gromacs.org/mailman/listinfo/gmx-users" target="_blank">http://lists.gromacs.org/mailman/listinfo/gmx-users</a><br> Please search the archive at<br> <a href="http://www.gromacs.org/Support/Mailing_Lists/Search" target="_blank">http://www.gromacs.org/Support/Mailing_Lists/Search</a> before posting!<br>
Please don't post (un)subscribe requests to the list. Use the<br> www interface or send it to <a href="mailto:gmx-users-request@gromacs.org" target="_blank">gmx-users-request@gromacs.org</a><br></div> <mailto:<a href="mailto:gmx-users-request@gromacs.org" target="_blank">gmx-users-request@gromacs.org</a>>.
<div class="im"><br> Can't post? Read <a href="http://www.gromacs.org/Support/Mailing_Lists" target="_blank">http://www.gromacs.org/Support/Mailing_Lists</a><br><br><br><br><br> -- Quaerendo Invenietis-Seek and you shall discover.<br>
<br><br><br><br>-- <br>Quaerendo Invenietis-Seek and you shall discover.<br><br></div></blockquote><br>-- <br>
<div class="im">========================================<br><br>Justin A. Lemkul<br>Ph.D. Candidate<br>ICTAS Doctoral Scholar<br>MILES-IGERT Trainee<br>Department of Biochemistry<br>Virginia Tech<br>Blacksburg, VA<br>jalemkul[at]<a href="http://vt.edu/" target="_blank">vt.edu</a> | (540) 231-9080<br>
<a href="http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin" target="_blank">http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin</a><br><br>========================================<br>-- <br></div>
<div>
<div></div>
<div class="h5">gmx-users mailing list <a href="mailto:gmx-users@gromacs.org" target="_blank">gmx-users@gromacs.org</a><br><a href="http://lists.gromacs.org/mailman/listinfo/gmx-users" target="_blank">http://lists.gromacs.org/mailman/listinfo/gmx-users</a><br>
Please search the archive at <a href="http://www.gromacs.org/Support/Mailing_Lists/Search" target="_blank">http://www.gromacs.org/Support/Mailing_Lists/Search</a> before posting!<br>Please don't post (un)subscribe requests to the list. Use the www interface or send it to <a href="mailto:gmx-users-request@gromacs.org" target="_blank">gmx-users-request@gromacs.org</a>.<br>
Can't post? Read <a href="http://www.gromacs.org/Support/Mailing_Lists" target="_blank">http://www.gromacs.org/Support/Mailing_Lists</a><br></div></div></blockquote></div><br><br clear="all"><br>-- <br>Quaerendo Invenietis-Seek and you shall discover.<br>