[gmx-users] Different results for same protein's simulation in 4.6.5 and 5.0.4

[Dries Van Rompaey]

Hi,
Looks like you might be applying positional restraints (define =-DPOSRES in
your mdp file). This is standard practice for equilibration, but I don't
think you want to be doing this for your production run, as this will
restrain your protein to conformations rather close to your starting
conformation. Check if this is the case in your logfile.

Proteins coming 'out of the box' is caused by periodic boundary conditions.
This is probably not a problem. Read this for more information:
http://www.gromacs.org/Documentation/Terminology/Periodic_Boundary_Conditions
Sometimes one pass of trjconv isn't enough to sort everything out and more
might be needed.

Kind regards

Dries
On 29 Feb 2016 5:25 a.m., "SAPNA BORAH" <sapnauser365 at gmail.com> wrote:

> Dear Mark,
>
> I have still not been able to solve my problem. A whole new problem has now
> arisen. I am listing them as follows:
>
> 1. I am running 100ns simulations for 6 protein models and I was using both
> version 4.6.5 and 5.0.4 when I found that after production run, my protein
> models are behaving differently in both the versions. I am attaching the
> images of rmsd differences in the link attached. Also, find the mdp files
> attached used in both the versions.
>
> 2. I have run multiple runs for the models to see if there are changes, but
> everytime, each of the models is very very stable and does not change in
> conformation at all in 5.0.4 but does otherwise in 4.6.5.
>
> 3. There is another problem that the protein comes out of the water box in
> 4.6.5 and not in 5.0.4 in all the simulations. I have ensured to remove
> jumps from the trajectories, make molecules whole.
>
> Please help me in this regard.
>
> http://dropcanvas.com/v0k0b
>
> Regards,
> Sapna.
>
> Sapna Mayuri Borah
> c/o Dr. A. N. Jha
> Research student
> Tezpur University,
> India
>
> On Thu, Feb 11, 2016 at 10:54 AM, Tsjerk Wassenaar <tsjerkw at gmail.com>
> wrote:
>
> > Hi Sapna,
> >
> > The list doesn't allow attachments. Please put them somewhere online and
> > provide a link.
> > The differences can be large. Can't tell if you don't know the possible
> > range of conformations/trajectories for one Gromacs version.
> > Did you ensure to remove jumps from the trajectories and/or make
> molecules
> > whole, etc? That would be the only thing we'd be able to see from the
> RMSD
> > plot anyway, whether you removed jumps or not. Otherwise it does not
> > contain useful information (in this regard).
> >
> > Cheers,
> >
> > Tsjerk
> > On Feb 11, 2016 05:28, "SAPNA BORAH" <sapnauser365 at gmail.com> wrote:
> >
> > > Dear Mark,
> > >
> > > Please find attached a plot showing the RMSD difference. It is for the
> > same
> > > protein.
> > >
> > > Even if the same runs are not reproducible, my doubt is whether there
> can
> > > be such a large variation in the RMSD trend. Which data should I rely
> on
> > > then, it is giving me a hard time.
> > >
> > > I'll be grateful if you look into this.
> > >
> > > Thanks,
> > > Sapna.
> > >
> > > Sapna Mayuri Borah
> > > c/o Dr. A. N. Jha
> > > Research student
> > > Tezpur University,
> > > India
> > >
> > > On Tue, Feb 9, 2016 at 1:07 PM, Mark Abraham <mark.j.abraham at gmail.com
> >
> > > wrote:
> > >
> > > > Hi,
> > > >
> > > > This is quite normal.
> > > > http://www.gromacs.org/Documentation/Terminology/Reproducibility.
> You
> > > > should see a similar range of variation in the trajectories produced
> in
> > > the
> > > > two versions that are incompletely sampling the same ensemble, but
> one
> > > can
> > > > say very little about two single observed trajectories sampled from
> > them.
> > > >
> > > > Mark
> > > >
> > > > On Tue, Feb 9, 2016 at 8:23 AM SAPNA BORAH <sapnauser365 at gmail.com>
> > > wrote:
> > > >
> > > > > Dear all,
> > > > >
> > > > > I have tried running the same protein in the two versions of
> gromacs
> > > > 4.6.5
> > > > > and 5.0.4. Now, both the runs are producing entirely different
> > results.
> > > > >
> > > > > The results till equilibration are same.
> > > > >
> > > > > After the run however, there is a total change in the production
> > > > results. I
> > > > > have used the same mdp files.
> > > > >
> > > > > I have rerun the systems and they still produce similar results in
> > the
> > > > same
> > > > > version.
> > > > >
> > > > > The protein structure is a modelled one.
> > > > >
> > > > >
> > > > > Thanks in advance.
> > > > > Sapna.
> > > > >
> > > > > Sapna Mayuri Borah
> > > > > c/o Dr. A. N. Jha
> > > > > Research student
> > > > > Tezpur University,
> > > > > India
> > > > > --
> > > > > Gromacs Users mailing list
> > > > >
> > > > > * Please search the archive at
> > > > > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> > > > > posting!
> > > > >
> > > > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> > > > >
> > > > > * For (un)subscribe requests visit
> > > > > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users
> > or
> > > > > send a mail to gmx-users-request at gromacs.org.
> > > > >
> > > > --
> > > > Gromacs Users mailing list
> > > >
> > > > * Please search the archive at
> > > > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> > > > posting!
> > > >
> > > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> > > >
> > > > * For (un)subscribe requests visit
> > > > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users
> or
> > > > send a mail to gmx-users-request at gromacs.org.
> > > >
> > >
> > > --
> > > Gromacs Users mailing list
> > >
> > > * Please search the archive at
> > > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> > > posting!
> > >
> > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> > >
> > > * For (un)subscribe requests visit
> > > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> > > send a mail to gmx-users-request at gromacs.org.
> > >
> > --
> > Gromacs Users mailing list
> >
> > * Please search the archive at
> > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> > posting!
> >
> > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> >
> > * For (un)subscribe requests visit
> > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> > send a mail to gmx-users-request at gromacs.org.
> >
> --
> Gromacs Users mailing list
>
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> posting!
>
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> send a mail to gmx-users-request at gromacs.org.
>