[gmx-users] g_current

[Florian Dommert]

Hi Nilesh,

 I looked at the code and was not able to find a reasonable condition, where
the last column of the mc output can contain -inf.
The only thing where the normalization would produce something like this is,
when the number of frames (nfr) equals the counter in the loop i for the
integral calculation. However this is a while loop with the condition
while(i<nfr) ( see line 291ff of gmx_current.c. Does this error always occur
or only for a certain trajectory? Do you use the whole system as index
group, which is necessary?

Unfortunately I have no trr lying around on my own computer. Can you provide
me access to a short trajectory, say a few thousand saved frames, including
the required top, mdp and gro files such that I can generate a corresponding
tpr?

/Flo

> -----Original Message-----
> From: gmx-users-bounces at gromacs.org [mailto:gmx-users-
> bounces at gromacs.org] On Behalf Of Nilesh Dhumal
> Sent: Sunday, September 08, 2013 11:55 PM
> To: Discussion list for GROMACS users
> Subject: RE: [gmx-users] g_current
> 
> Did you get chance to take look at the g_current?
> 
> why it is     -inf  (infinity)  at the end with -mc option (mc.xvg).
> >
> >
> >> -----Original Message-----
> >> From: gmx-users-bounces at gromacs.org [mailto:gmx-users-
> >> bounces at gromacs.org] On Behalf Of Nilesh Dhumal
> >> Sent: Tuesday, July 09, 2013 5:06 AM
> >> To: Discussion list for GROMACS users
> >> Subject: RE: [gmx-users] g_current
> >>
> >> Thanks for reply.
> >>
> >> Still I have a question why it is     -inf  (infinity)  at the end.
> >
> > This is a good question. At the moment, the only thing I can imagine is
a
> > faulty normalization, because the value of the integral at t=0 equals
> > always
> > zero. However, if you integrate the first row with another tool, such as
> > g_analyze for example, the same values as in the second row show appear.
> >
> > As soon as I have some time left, I will take a look at the code and
> > report
> > back.
> >
> > /Flo
> >
> >>
> >> Nilesh
> >>
> >> > Hi,
> >> >
> >> >  I would say the problem is too less statistics. In the paper where
> >> the
> >> > tool
> >> > was introduced, a simulation of 100ns has been performed in order to
> >> > achieved sufficient sampling.
> >> >
> >> > The order of magnitude is given due to the units.
> >> >
> >> > Simulate for a longer time to get rid of the noise. Calculating these
> > kind
> >> > of correlations functions is really not a simple task.
> >> >
> >> > Schroeder et al. have published an article about fitting of
dielectric
> >> > spectra for ionic liquids, where this problem is discussed in detail.
> >> >
> >> > Cheers,
> >> > Flo
> >> >
> >> >> -----Original Message-----
> >> >> From: gmx-users-bounces at gromacs.org [mailto:gmx-users-
> >> >> bounces at gromacs.org] On Behalf Of Nilesh Dhumal
> >> >> Sent: Monday, July 08, 2013 10:32 PM
> >> >> To: gmx-users at gromacs.org
> >> >> Subject: [gmx-users] g_current
> >> >>
> >> >> Hello,
> >> >> I am calculating the correlation of rotational and translation
dipole
> >> >> moment of ionic liquids. I run the 1 ns simulation and saved the
> >> >> trajectory at 1 fs.
> >> >>
> >> >> g_current -f md.trr -s md.tpr -n index.ndx -mc
> >> >>
> >> >> I selected "system" group.
> >> >>
> >> >> Here I pasted the initial j(t) vales and the final value is
infinity.
> >> >>
> >> >> # This file was created Sun Jul  7 20:51:59 2013
> >> >> # by the following command:
> >> >> # g_current -f md.trr -s md.tpr -e 10.0 -mc
> >> >> #
> >> >> # g_current is part of G R O M A C S:
> >> >> #
> >> >> # Gallium Rubidium Oxygen Manganese Argon Carbon Silicon
> >> >> #
> >> >> @    title "M\sD\N - current  autocorrelation function"
> >> >> @    xaxis  label "Time (ps)"
> >> >> @    yaxis  label "< M\sD\N (0)\c7\CJ(t) >  (e nm/ps)\S2"
> >> >> @TYPE xy
> >> >> # time   M_D(0) J(t) acf         Integral acf
> >> >> 0.000   -5.20584e-15             0
> >> >> 0.001   -6.97466e-15    -1.42702e-17
> >> >> 0.002   -6.97511e-15    -5.25081e-17
> >> >> 0.003   -6.97474e-15    -9.07484e-17
> >> >> 0.004   -6.97391e-15    -1.28987e-16
> >> >>
> >> >>
> >> >>
> >> >> Why final value is infinity? If I do autocorrelation function of
> >> these
> >> >> values, I get all values 0.
> >> >>
> >> >> Why J(t)values are in range of e-15?
> >> >>
> >> >> Nilesh
> >> >>
> >> >>
> >> >>
> >> >> --
> >> >> gmx-users mailing list    gmx-users at 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-request at gromacs.org.
> >> >> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> >> >
> >> > --
> >> > gmx-users mailing list    gmx-users at 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-request at gromacs.org.
> >> > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> >> >
> >>
> >>
> >> --
> >> gmx-users mailing list    gmx-users at 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-request at gromacs.org.
> >> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> >
> > --
> > gmx-users mailing list    gmx-users at 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-request at gromacs.org.
> > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> >
> 
> --
> gmx-users mailing list    gmx-users at 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-request at gromacs.org.
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists