Hi to everybody, Bharat, maybe i didn't follow exactly the wole tale, but is it possible you are running xmgrace without the -nxy option? You are probably visualizing the data related the 1st replica several times!
Francesco 2013/5/24 Mark Abraham <mark.j.abra...@gmail.com> > On Fri, May 24, 2013 at 10:44 AM, bharat gupta <bharat.85.m...@gmail.com > >wrote: > > > Dear Sir, > > > > Thank you for your detailed response to my query. I understood the > concept > > of ordered arrangement of ensembles in replica_index.xvg. But I have a > > doubt, you said that " *At time 4, replicas in ensemble 1 and 2 have > > exchanged. So replica 0 is now in ensemble 2, which is expressed by 0 in > > the third column* > > *of the third row of replica_index.xvg.* " > > > > This is fine , as the output of replica_index is :- > > 4 1 2 *0 * 3 4 5 6 7 8 9 10 11 > 13 > > 12 > > > > But, i didn't understand this " The same condition is expressed by the > > first column of the third row of replica_temp.xvg, where you will find 2, > > also expressing that replica 0 is in ensemble 2 at time 4." Here's the > > output for replica_temp . The first column third row is 2, its ok, but, > its > > shows that replica 0 is in ensemble 1 instead of 2. > > > > No, if the rows of both matrices describe time, and there are two different > matrices for the same exchange set, then the information described by a > column must differ, like I said last email. You are applying the same > interpretation to a column from either matrix. > > > > 4 2 0 1 3 4 5 6 7 8 9 10 11 13 > > 12 > > > > > > In addition to this, in my last mail I showed the temp graph for all > > replicas. (https://www.dropbox.com/s/0c8gp584v1hvlbx/replica_temp.png) . > > Not all replicas visit all the temperatures, but some of them visit all > the > > temperatures. Is it sufficient to move with the further analysis , as in > > some papers they show that some replicas visit all the temp which means > > that the sufficient sampling has been achieved. In my case this is true > for > > some of the replicas and the average acceptance ratio achieved was 0.22 > ?? > > > > I've answered this question several times. Each replica merely visiting > each temperature means nothing for converged sampling. There's lots of > literature here, including stuff by me ;-) A balance of replicas visiting > ensembles is necessary but not sufficient for the kind of replica flow that > would be necessary for generalized convergence. One can shrug one's > shoulders at some point and say things are probably as good as they'll get > for reasonable cost, but your reviewer might disagree with you. Convergence > of sampling at a single temperature can be assessed in a similar way as for > non-REMD simulations, caveat that the exchange events pretty much stop you > using metrics based on correlation time. If you want to know how to do > things properly, you need to do some reading. > > Mark > > > > > > > > > > On Fri, May 24, 2013 at 5:07 PM, Mark Abraham <mark.j.abra...@gmail.com > > >wrote: > > > > > At time 0 we have an set of replicas and an (ordered) set of ensembles. > > We > > > could label these however we liked, but for (in)convenience we use > > 0-(n-1) > > > for both. The rows of the matrices in the *.xvg files change with time. > > At > > > time 2, replicas in ensemble 0 and 1 have exchanged, so replica 0 is > now > > in > > > ensemble 1. At time 4, replicas in ensemble 1 and 2 have exchanged. So > > > replica 0 is now in ensemble 2, which is expressed by 0 in the third > > column > > > of the third row of replica_index.xvg. The same condition is expressed > by > > > the first column of the third row of replica_temp.xvg, where you will > > find > > > 2, also expressing that replica 0 is in ensemble 2 at time 4. The > columns > > > of the two matrices allow you to see either the profile of which > replica > > > was in this ensemble at which time, or which ensemble this replica was > in > > > at which time. > > > > > > Mark > > > > > > > > > On Fri, May 24, 2013 at 8:43 AM, bharat gupta < > bharat.85.m...@gmail.com > > > >wrote: > > > > > > > Dear Sir, > > > > > > > > I tried a lot to understand the meaning and relation between the .log > > > file > > > > and relica_index file, but I was not able to break the code. I tried > to > > > > look into gmx forum for some clue, but didn't find any. So, if > possible > > > can > > > > you explain it ... > > > > > > > > Replica exchange at step 1000 time 2 > > > > Repl 0 <-> 1 dE = -1.067e+00 > > > > Repl ex 0 x 1 2 3 4 5 6 7 8 9 10 11 > 12 > > x > > > 13 > > > > Repl pr 1.0 .01 .68 .21 .05 .09 > > .26 > > > > > > > > > > > > Replica exchange at step 2000 time 4 > > > > Repl ex 0 1 x 2 3 4 5 6 7 8 9 10 11 > 12 > > > 13 > > > > Repl pr .91 .32 .00 .07 .18 .08 > > > > > > > > output of replica_index.xvg > > > > > > > > 0 0 1 2 3 4 5 6 7 8 9 10 11 > > 12 > > > > 13 > > > > 2 1 0 2 3 4 5 6 7 8 9 10 11 > > 13 > > > > 12 > > > > 4 1 2 0 3 4 5 6 7 8 9 10 11 > > 13 > > > > 12 > > > > > > > > On Thu, May 23, 2013 at 11:20 PM, Mark Abraham < > > mark.j.abra...@gmail.com > > > > >wrote: > > > > > > > > > Looked fine > > > > > > > > > > > > > > > On Thu, May 23, 2013 at 4:13 PM, bharat gupta < > > > bharat.85.m...@gmail.com > > > > > >wrote: > > > > > > > > > > > Sir, > > > > > > > > > > > > What about the description of replica_temp file that I posted in > > last > > > > > mail. > > > > > > I think that's correct ... If you can comment on that, I can move > > on > > > > with > > > > > > replica_index file... > > > > > > > > > > > > > > > > > > On Thu, May 23, 2013 at 10:58 PM, Mark Abraham < > > > > mark.j.abra...@gmail.com > > > > > > >wrote: > > > > > > > > > > > > > It's a demux. One might want trajectories to be at constant > > > > > temperature, > > > > > > or > > > > > > > constant replica. The two files define the (mutually inverse) > > > > mappings > > > > > > > between those representations. So one file tells you which > > replica > > > is > > > > > at > > > > > > > each temperature, and the other which temperature holds each > > > replica. > > > > > > > Nobody's ever written down anything about which is which, so > > like I > > > > > said > > > > > > a > > > > > > > week back, look at the first few exchanges, see how those are > > > > > represented > > > > > > > in the files, and decide for yourself which file's columns/rows > > > have > > > > > > useful > > > > > > > information you want to look at. And do write that decision > down! > > > :-) > > > > > > > > > > > > > > Mark > > > > > > > > > > > > > > > > > > > > > On Thu, May 23, 2013 at 2:55 PM, simula_460 < > > > > bharat.85.m...@gmail.com > > > > > > > >wrote: > > > > > > > > > > > > > > > I checked the md.log and replica_temp.xvg file , what I > > > understood > > > > is > > > > > > > that > > > > > > > > the 'x' means swapping and replica are written this way. > > > > > > > > For eg. > > > > > > > > Replica exchange at step 1000 time 2 > > > > > > > > Repl 0 <-> 1 dE = -1.067e+00 > > > > > > > > Repl ex 0 x 1 2 3 4 5 6 7 8 9 10 > > 11 > > > > > 12 > > > > > > x > > > > > > > 13 > > > > > > > > Repl pr 1.0 .01 .68 .21 .05 > .09 > > > > > > .26 > > > > > > > > > > > > > > > > output in replica_temp file will be > > > > > > > > 1 0 2 3 4 5 6 7 8 9 10 11 13 12 > > > > > > > > > > > > > > > > It means that replica 1 at higher temp. exchange with the one > > in > > > > > lower > > > > > > > temp > > > > > > > > 0. > > > > > > > > > > > > > > > > Replica exchange at step 2000 time 4 > > > > > > > > Repl ex 0 1 x 2 3 4 5 6 7 8 9 10 > > 11 > > > > > 12 > > > > > > > 13 > > > > > > > > Repl pr .91 .32 .00 .07 .18 > > > > .08 > > > > > > > > > > > > > > > > output in replica_temp file will be > > > > > > > > 1 0 2 3 4 5 6 7 8 9 10 11 13 12 > > > > > > > > 2 0 1 3 4 5 6 7 8 9 10 11 13 12 > > > > > > > > > > > > > > > > [order is from low to high temp] > > > > > > > > > > > > > > > > But I am not able to understand for replica_index file :- > > > > > > > > > > > > > > > > for the above two time steps here's the output :- > > > > > > > > 0 0 1 2 3 4 5 6 7 8 9 > 10 > > > > 11 > > > > > > 12 > > > > > > > > 13 > > > > > > > > 2 1 0 2 3 4 5 6 7 8 9 > 10 > > > > 11 > > > > > > 13 > > > > > > > > 12 > > > > > > > > 4 1 2 0 3 4 5 6 7 8 9 > 10 > > > > 11 > > > > > > 13 > > > > > > > > 12 > > > > > > > > > > > > > > > > The time step four is different here, I don't know why ?? > > Ideally > > > > the > > > > > > > > output should be same in both files, I suppose ?? > > > > > > > > > > > > > > > > Also, I tried to plot for each column separately , here I > want > > to > > > > > > clarify > > > > > > > > that whether each column represents the time evolution of > each > > > > > replica > > > > > > > over > > > > > > > > time. For eg. the second column should represent the temp > > > evolution > > > > > for > > > > > > > > replica No. 0 wrt to time. Presuming that I understood it > > > > correctly, > > > > > I > > > > > > > > plotted the temp. evolution over time of all replicas > > separately > > > . > > > > > > Here's > > > > > > > > the replica_temp plot for replicas 0 to 13. > > > > > > > > > > > > > > > > https://www.dropbox.com/s/0c8gp584v1hvlbx/replica_temp.png > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > -- > > > > > > > > View this message in context: > > > > > > > > > > > > > > > http://gromacs.5086.x6.nabble.com/REMD-analysis-tp5008199p5008481.html > > > > > > > > Sent from the GROMACS Users Forum mailing list archive at > > > > Nabble.com. > > > > > > > > -- > > > > > > > > gmx-users mailing list gmx-users@gromacs.org > > > > > > > > http://lists.gromacs.org/mailman/listinfo/gmx-users > > > > > > > > * Please search the archive at > > > > > > > > http://www.gromacs.org/Support/Mailing_Lists/Search before > > > > posting! > > > > > > > > * Please don't post (un)subscribe requests to the list. Use > the > > > > > > > > www interface or send it to gmx-users-requ...@gromacs.org. > > > > > > > > * Can't post? Read > > http://www.gromacs.org/Support/Mailing_Lists > > > > > > > > > > > > > > > -- > > > > > > > gmx-users mailing list gmx-users@gromacs.org > > > > > > > http://lists.gromacs.org/mailman/listinfo/gmx-users > > > > > > > * Please search the archive at > > > > > > > http://www.gromacs.org/Support/Mailing_Lists/Search before > > > posting! > > > > > > > * Please don't post (un)subscribe requests to the list. Use the > > > > > > > www interface or send it to gmx-users-requ...@gromacs.org. > > > > > > > * Can't post? Read > http://www.gromacs.org/Support/Mailing_Lists > > > > > > > > > > > > > -- > > > > > > gmx-users mailing list gmx-users@gromacs.org > > > > > > http://lists.gromacs.org/mailman/listinfo/gmx-users > > > > > > * Please search the archive at > > > > > > http://www.gromacs.org/Support/Mailing_Lists/Search before > > posting! > > > > > > * Please don't post (un)subscribe requests to the list. Use the > > > > > > www interface or send it to gmx-users-requ...@gromacs.org. > > > > > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists > > > > > > > > > > > -- > > > > > gmx-users mailing list gmx-users@gromacs.org > > > > > http://lists.gromacs.org/mailman/listinfo/gmx-users > > > > > * Please search the archive at > > > > > http://www.gromacs.org/Support/Mailing_Lists/Search before > posting! > > > > > * Please don't post (un)subscribe requests to the list. Use the > > > > > www interface or send it to gmx-users-requ...@gromacs.org. > > > > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists > > > > > > > > > -- > > > > gmx-users mailing list gmx-users@gromacs.org > > > > http://lists.gromacs.org/mailman/listinfo/gmx-users > > > > * Please search the archive at > > > > http://www.gromacs.org/Support/Mailing_Lists/Search before posting! > > > > * Please don't post (un)subscribe requests to the list. Use the > > > > www interface or send it to gmx-users-requ...@gromacs.org. > > > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists > > > > > > > -- > > > gmx-users mailing list gmx-users@gromacs.org > > > http://lists.gromacs.org/mailman/listinfo/gmx-users > > > * Please search the archive at > > > http://www.gromacs.org/Support/Mailing_Lists/Search before posting! > > > * Please don't post (un)subscribe requests to the list. Use the > > > www interface or send it to gmx-users-requ...@gromacs.org. > > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists > > > > > -- > > gmx-users mailing list gmx-users@gromacs.org > > http://lists.gromacs.org/mailman/listinfo/gmx-users > > * Please search the archive at > > http://www.gromacs.org/Support/Mailing_Lists/Search before posting! > > * Please don't post (un)subscribe requests to the list. Use the > > www interface or send it to gmx-users-requ...@gromacs.org. > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists > > > -- > gmx-users mailing list gmx-users@gromacs.org > http://lists.gromacs.org/mailman/listinfo/gmx-users > * Please search the archive at > http://www.gromacs.org/Support/Mailing_Lists/Search before posting! > * Please don't post (un)subscribe requests to the list. Use the > www interface or send it to gmx-users-requ...@gromacs.org. > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists > -- Cordiali saluti, Dr.Oteri Francesco -- gmx-users mailing list gmx-users@gromacs.org http://lists.gromacs.org/mailman/listinfo/gmx-users * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/Search before posting! * Please don't post (un)subscribe requests to the list. Use the www interface or send it to gmx-users-requ...@gromacs.org. * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
