Okay, now I can start with large production runs .
On Thu, May 16, 2013 at 11:10 PM, XAvier Periole <x.peri...@rug.nl> wrote: > > Indeed the Repeat-3 seems good. But I would guess you did not run too > long, right! That would explain the distribution of values! > > On May 16, 2013, at 2:04 PM, bharat gupta <bharat.85.m...@gmail.com> > wrote: > > > 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 > > -- > 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
