[gmx-users] ED analysis

[Bert de Groot]

vincenzo venditti wrote:
> Hi all,
> I'm trying to carry out a combined analysis of essential dynamics of two 
> similar systems (one with a ligand and the other without it).
> I've fitted the trajectories to the same reference structure using trjconv.
> Then I've concatenated the fitted trajectories using trjcat command.
> I've constructed and diagonalized the Cá covariance matrix using g_covar 
> command and an index file in which only Cá atoms of protein reidues are 
> considered.
> I've calculated averaged values of the projections of the separate 
> trajectories onto the concatenated eigenvectors using the -proj option 
> of g_anaeig command and averaging out the projections thus obtained for 
> each eigenvector, but the results have been
> 
> eigenvec 1:      -3.17    -1.84
> eigenvec 2:    -10.21    -9.06
> eigenvec 3:      1.96      3.70
> 
> (Where the first column is the protein with the ligand, and the 
> secondcolumn is the protein without ligand)
> The two individual simulations have the same length.
> Why have I not obteined simmetrical results with respect to zero?
> 

this is indeed strange. This means that the average projection averaged over
the complete, concatenated trajectory is not zero? This should usually not
happen, unless you use non-default options in g_covar for -fit and -ref, or
if you use different trajectories or reference structures for g_covar and g_anaeig.


Bert

____________________________________________________________________________
Dr. Bert de Groot

Max Planck Institute for Biophysical Chemistry
Computational biomolecular dynamics group
Am Fassberg 11
37077 Goettingen, Germany

tel: +49-551-2012308, fax: +49-551-2012302

email: bgroot at gwdg.de
http://www.mpibpc.gwdg.de/abteilungen/073
____________________________________________________________________________