Okay Sir, I will try two-three combinations this time and will report back to you ...
On Thu, May 16, 2013 at 5:25 PM, XAvier Periole <x.peri...@rug.nl> wrote: > > An acceptance ratio of 0.2/0.3 is normally best. The problem with high > acceptance ratio is that it means that a large portion of the exchanges are > just back and forth exchanges between consecutive exchange and are thus > disturbing the system more that actually helping sampling. > > I do not know particularly the paper you mention but if you like what they > do, it is your choice at the end. > > Why don;t you just increase the spacing between the replicas? You will > need less replicas and potentially you could run two simulations instead of > one and evaluate the convergence ... > > On May 16, 2013, at 1:50 AM, bharat gupta <bharat.85.m...@gmail.com> > wrote: > > > The plots that I showed in my last mail were for all replicas. I tried > > plotting the first 500 ps of replica_index and replica_time files. I > think > > the plots look fine, and there could be problem with the plotting tool . > > Here the link for both files , > > https://www.dropbox.com/s/2g16mlxfsme4rx2/replica_temp.bmp > > https://www.dropbox.com/s/8jfs0b9whu6j7lo/replica_index.bmp > > > > Now regarding the high acceptance ratio which is 0.5 , I came across a > > paper (http://www.pnas.org/content/100/13/7587.full.pdf), here they have > > mentioned that their average acceptance ratio ranged between 30 to 80%. I > > have a question here, how did they calculate the range for the average > > acceptance ratio or is it average ratio for each replica . Actually, this > > is the reference I am following. I am also interested in peptide folding > > simulation, similar to this article. > > > > I want to know, whether the average acceptance ratio that I have got for > my > > trial simulation is correct , together with the replica_temp and > > replica_remd plots. Can I proceed for large production runs to complete > my > > experiment ?? > > > > > > > > > > On Tue, May 14, 2013 at 6:34 PM, XAvier Periole <x.peri...@rug.nl> > wrote: > > > >> > >> The interval between the exchange trial affect the efficiency of REMD > but > >> not the the exchange ratio (at least in principle). > >> > >> In you case I am not sure what the plot are showing! Are these showing > all > >> the replicas? what are the units? > >> > >> On May 14, 2013, at 5:07 AM, bharat gupta <bharat.85.m...@gmail.com> > >> wrote: > >> > >>> Dear Sir, > >>> > >>> Here's the result for the REMD trial with large temperature gaps. > >>> > >>> Temp. distribution : 280.0 294.9 310.7 327.3 344.7 363.1 382.5 402.9 > >> 424.4 > >>> 447.1 471.0 496.1 522.6 550.5 579.9 610.8 > >>> > >>> Out of md16.log : > >>> > >>> Replica exchange statistics > >>> Repl 249 attempts, 125 odd, 124 even > >>> Repl average probabilities: > >>> Repl 0 1 2 3 4 5 6 7 8 9 10 11 12 > >>> 13 14 15 > >>> Repl .40 .34 .38 .43 .43 .36 .45 .40 .37 .48 .47 .45 > .47 > >>> .44 .46 > >>> > >>> Repl number of exchanges: > >>> Repl 0 1 2 3 4 5 6 7 8 9 10 11 12 > >>> 13 14 15 > >>> Repl 50 42 46 52 57 40 58 49 42 53 61 63 > 56 > >>> 57 58 > >>> > >>> Repl average number of exchanges: > >>> Repl 0 1 2 3 4 5 6 7 8 9 10 11 12 > >>> 13 14 15 > >>> Repl .40 .34 .37 .42 .46 .32 .46 .40 .34 .43 .49 .51 > .45 > >>> .46 .46 > >>> Average acceptance ratio : 0.46 > >>> > >>> But, the repli_index.xvg and replica_temp.xvg files still shows that > the > >>> replicas does not exchange equally well . > >>> > >>> https://www.dropbox.com/s/zkbwpuj7l2o282b/replica_index.png > >>> https://www.dropbox.com/s/0c8gp584v1hvlbx/replica_temp.png > >>> > >>> what could be wrong in this case?? Is it the mdp file settings or > >> implicit > >>> solvent setting. Does the time to replica to exhange also affects their > >>> swapping ?? > >>> > >>> > >>> > >>> On Tue, May 14, 2013 at 12:24 AM, XAvier Periole <x.peri...@rug.nl> > >> wrote: > >>> > >>>> > >>>> You need to increase the temperature gaps indeed if you want > acceptance > >>>> ratio ~0.2/0.3. But again this won't work with the water … > >>>> > >>>> It is not clear what happens in your index file but probably a problem > >>>> from grace to plot so many points … you can try to increase the "Max > >>>> drawing path length" in the preference menu of grace. > >>>> > >>>> On May 13, 2013, at 4:22 PM, bharat gupta <bharat.85.m...@gmail.com> > >>>> wrote: > >>>> > >>>>> Dear Sir, > >>>>> > >>>>> I repeated the simulation again for 25 replicas with the following > >> temp. > >>>>> distribution . > >>>>> > >>>>> 280 > >>>>> 289.1 > >>>>> 298.5 > >>>>> 308.2 > >>>>> 318.2 > >>>>> 328.6 > >>>>> 339.3 > >>>>> 350.3 > >>>>> 361.7 > >>>>> 373.5 > >>>>> 385.6 > >>>>> 398.1 > >>>>> 411.1 > >>>>> 424.4 > >>>>> 438.3 > >>>>> 452.5 > >>>>> 467.2 > >>>>> 482.4 > >>>>> 498.1 > >>>>> 514.3 > >>>>> 531.0 > >>>>> 548.3 > >>>>> 566.1 > >>>>> 584.5 > >>>>> 603.5 > >>>>> 623.2 > >>>>> > >>>>> The output of md.log file is :- > >>>>> > >>>>> Replica exchange statistics > >>>>> Repl 24999 attempts, 12500 odd, 12499 even > >>>>> Repl average probabilities: > >>>>> Repl 0 1 2 3 4 5 6 7 8 9 10 11 > 12 > >>>>> 13 14 15 16 17 18 19 20 21 22 23 24 25 > >>>>> Repl .63 .63 .62 .62 .61 .61 .60 .60 .59 .59 .58 .59 > >> .59 > >>>>> .60 .60 .61 .62 .62 .63 .64 .64 .65 .65 .66 .66 > >>>>> > >>>>> Repl number of exchanges: > >>>>> Repl 0 1 2 3 4 5 6 7 8 9 10 11 > 12 > >>>>> 13 14 15 16 17 18 19 20 21 22 23 24 25 > >>>>> Repl 7822 7752 7816 7760 7639 7628 7511 7442 7375 7332 7312 7424 > >> 7408 > >>>>> 7410 7522 7559 7684 7697 7878 7927 7917 8073 8151 8208 8266 > >>>>> > >>>>> Repl average number of exchanges: > >>>>> Repl 0 1 2 3 4 5 6 7 8 9 10 11 > 12 > >>>>> 13 14 15 16 17 18 19 20 21 22 23 24 25 > >>>>> Repl .63 .62 .63 .62 .61 .61 .60 .60 .59 .59 .58 .59 > >> .59 > >>>>> .59 .60 .60 .61 .62 .63 .63 .63 .65 .65 .66 .66 > >>>>> > >>>>> The average acceptance ration is around 0.6 which is still high. > >>>>> > >>>>> The link for replica_temp,replica_index : > >>>>> https://www.dropbox.com/s/c7soajnwc3uww8j/replica_temp.png > >>>>> https://www.dropbox.com/s/wvx82m4c6cnsfit/replica_index.png > >>>>> > >>>>> The temp files look better but the index file looks weird ... > >>>>> > >>>>> Do i need to experiment with the gap difference in order to get the > >>>>> required ration of 0.2-0.3 ?? There is some problem with the .mdp > file > >>>>> settings?? > >>>>> > >>>>> -- > >>>>> Bharat > >>>>> -- > >>>>> 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