[gmx-users] How to use different reference structures for g_covar and g_anaeig

[Tsjerk Wassenaar]

Hi Shridhar,

g_covar does not use the average structure for fitting. It uses the
structure supplied with -s for fitting and from the fitted trajectory
calculates the average structure to determine the deviations from.
First calculating the average structure and use that for fitting is
meaningless (unless you have fitted the trajectory onto some reference
before).

David is mistaken, since eigenvec.trr DOES contain the average
structure, which is stored in the frame with timestamp t=0. If a
reference was given, this too is stored in the .trr file as the frame
with time stamp t=-1 (this is in the manual afaik).

Now, your eigenvectors are defined for the deviations from the average
structure from your trajectory after fitting to the reference used in
g_covar. Refitting onto a different reference and projecting onto the
eigenvectors obtained will lead to meaningless results.

The mean square fluctuation of the projection of a trajectory onto an
eigenvector is the average deviation of each frame from its average
projection. You can certainly get it from the projection of each
trajectory separately (-proj), although there may be an option to do
it directly, or you could use g_analyze (not sure about that really).

I think it would be good if you read some general statistics texts
about principal component analysis to understand what it does and how
it would relate to fitting and average structures.

Hope it helps,

Tsjerk

On 5/16/07, David van der Spoel <spoel at xray.bmc.uu.se> wrote:
> sridhar at cdfd.org.in wrote:
> > Dear All,
> > I have two specific questions:
> >
> > 1.I have been to do COMBINED ESSENTIAL dynamics by concatenating two
> > trajectories. I calcualted the covariance matrix using the default options
> > of g_covar. This should presumably use the average structure for fitting.
> > Now I want to use one of a different structure other than the average
> > structure of the two trajectories in g_anaeig. It was said in the
> > gmx-mailing lists that g_anaeig uses the structure in the eigenvec.trr
> > file if present, otherwise it takes the structure given in the -s option.
> > But I suppose the structure file is always present in the eigenvec.trr
> > file. Is there any way that we can ignore the structure file in
> > eigenvec.trr file and use the file supplied in -s option of g_anaeig?
> the trr file does not contain a structure, but it one can deduce an
> average from it.
> >
> > 2. Further, in the case of COMBINED ESSENTIAL DYNAMICS, many researchers
> > could compute the Mean square fluctuation by individual trajectories
> > projected on to the combined eigenvectors using the covariance matrix. Can
> > any one explain how to calculate this?
> probably it is done separately as well as combined.
>
> > g_anaeig of COMBINED ESSENTIAL DYNAMICS can give eigenrmsf of only the
> > combined trajectory!
> that is correct I think.
> >
> > Thanks
> >
> > _______________________________________________
> > gmx-users mailing list    gmx-users at gromacs.org
> > http://www.gromacs.org/mailman/listinfo/gmx-users
> > Please search the archive at http://www.gromacs.org/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/mailing_lists/users.php
>
>
> --
> David.
> ________________________________________________________________________
> David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics group,
> Dept. of Cell and Molecular Biology, Uppsala University.
> Husargatan 3, Box 596,          75124 Uppsala, Sweden
> phone:  46 18 471 4205          fax: 46 18 511 755
> spoel at xray.bmc.uu.se    spoel at gromacs.org   http://folding.bmc.uu.se
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at http://www.gromacs.org/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/mailing_lists/users.php
>


-- 
Tsjerk A. Wassenaar, Ph.D.
Junior UD (post-doc)
Biomolecular NMR, Bijvoet Center
Utrecht University
Padualaan 8
3584 CH Utrecht
The Netherlands
P: +31-30-2539931
F: +31-30-2537623