Dear Sir, Here's the result of three different runs :
Temperature distribution for three trials Repeat-1 280 298 317 337 359 382 406 432 460 489 520 554 589 627 Repeat-2 280 299 319 340 363 388 414 441 471 503 536 572 611 Repeat-3 280 300 322 345 370 397 426 457 490 526 564 605 649 md.log files output from three different trials: Repeat-1 .37 .28 .26 .30 .25 .29 .32 .35 .32 .35 .36 .32 .31 Repeat-2 .30 .33 .30 .25 .19 .27 .30 .31 .27 .40 .34 .31 Repeat-3 .18 .22 .26 .34 .26 .28 .25 .27 .27 .25 .27 .22 I think as the required acceptance value all the three trials are fine, but trail 3 would be much better to continue the further runs and anlysis ?? So, is it fine to continue with the third simulation ?? But still the problem is that I am not getting the exact graphs with xmgrace?? On Thu, May 16, 2013 at 5:36 PM, XAvier Periole <x.peri...@rug.nl> wrote: > > You have to convince yourself, not me :)) But I can give you my opinion … > > On May 16, 2013, at 10:33 AM, bharat gupta <bharat.85.m...@gmail.com> > wrote: > > > 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 > > -- > 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
