[gmx-users] Combined essential dynamics analysis

[Bert de Groot]

On Wed, 16 Jul 2003, Ruben Martinez Buey wrote:

> Hi all,
> I´m trying to compare two similar systems (one with a ligand and the other
> without it). I´ve used CONCOORD for both systems. I´ve concatenated the two
> individual trajectories and run g_covar over the concatenated trajectory. Only
> the first 5-7 resulting combined eigenvectors are significant.
> When I project the individual trajectories over these combined eigenvectors, I
> get the next results:
>
> Average projection (mean):
> eigenvec 1: -0.365220            0.457450
> eigenvec 2:   0.137630          -0.169460
> eigenvec 3:   0.137790          -0.319940
> eigenvec 4:   0.068647          -0.078291
> eigenvec 5:   0.087228            0.040858

hmm, I suppose the two trajectories had a different length? Since
otherwise the projections should be symmetrical with respect to zero.
( ie -0.3   0.3 etc)


>
> Mean square fluctations (stddev):
> eigenvec 1:   6.7204                 6.7390
> eigenvec 2:   3.9814                 4.2909
> eigenvec 3:   3.6093                 3.6686
> eigenvec 4:   2.3390                 2.2201
> eigenvec 5:   1.8614                 1.8681
>
> (Where the first column is the protein with the ligand, and the second column
> is the protein without ligand)
> But the extreme projections over each eigenvector (generated by g_anaeig) are
> exactly the same in both cases. So, what would be the conclusion? There are no
> significant differences in the essential dynamics beetwen both systems? But the
> differences in the average projections are large enough to be significant,
> aren´t they? And the
>

right. although the differences in the averages are small as compared to
the mean square fluctuations, they *may* still be significant.
Look to what kind of motion the first eigenvector corresponds to (ie if eg
a negative projection corresponds to a more closed conformation or
whatever) and then see what that implies for the effect of the ligand
binding to the protein conformation. and repeat this for ev 2 and 3
(the others are certainly not significant)

Bert

____________________________________________________________________________
Dr. Bert de Groot

Max Planck Institute for Biophysical Chemistry
Theoretical molecular biophysics group
Am Fassberg 11
37077 Goettingen, Germany

tel: +49-551-2011306, fax: +49-551-2011089

email: bgroot at gwdg.de
http://www.mpibpc.gwdg.de/abteilungen/071/bgroot
____________________________________________________________________________