[gmx-users] help on g_cluster transitions

[Ran Friedman]

Dear Caterina, GMX users,

Whether or not the various clusters really represented different
conformational states depends on your system and on the definition of
the clusters (and of the conformational state). You can have some idea
about the difference between the clusters by examining the central
structures. However, a single measure (e. g. clustering analysis) is
generally not enough to gain structural insights in complex structures
such as proteins.

The same is true for stability. The fact that you have transitions
cannot state whether or not the simulation is stable (unless you
simulate only for a short time). You may want to check other parameters
such as the overall and matrix rmsd, secondary structure, energy
component etc. to see if the simulation is stable. Examining the
trajectory can also help.

Hope it helped,
Ran.

Caterina Arcangeli wrote:
> Hi all,
> I'm doing cluster analysis of my protein simulation.
> I use gromos method on the Ca atoms and I obtain 10 clusters.
>
> Is not clear to me the (biological) meaning of transition between
> clusters.
>
> 1. Are transitions occurring among different conformational states?
> 2. What does means if one found several clusters-transitions: maybe
> that the simulation is not stable?
>
> Thank to all.
> Caterina
>
>
> P.S.
> Here is the .log file I've obtained:
>
> ------
> Using gromos method for clustering
> Using RMSD cutoff 0.1 nm
> The RMSD ranges from 0.046404 to 0.181787 nm
> Average RMSD is 0.117163
> Number of structures for matrix 561
> Energy of the matrix is 23.5304 nm
>
> Found 10 clusters
>
> Writing average structure for each cluster to
> cluster/pdb_average/F8M47_cluster.pdb
> Writing all structures for clusters with more than 1 structures to
> cluster/pdb_average/F8M47_cluster%02d.pdb
> Counted 106 transitions in total, max 17 between two specific clusters
>
>
> cl. | #st rmsd | middle rmsd | cluster members
>   1 | 264 .103 |   2832 .090 |   2302   2324   2370   2372   2374   2380
>   2 | 128 .097 |   2082 .086 |   2020   2022   2024   2026   2028   2030
>   3 |  49 .099 |   2368 .090 |   2262   2268   2274   2276   2282   2298
>   4 |  48 .095 |   2650 .085 |   2514   2516   2518   2520   2526   2534
>   5 |  34 .088 |   3086 .077 |   3028   3032   3040   3048   3050   3052
>   6 |  18 .094 |   2260 .085 |   2232   2236   2238   2240   2242
> 2244    7 |  13 .092 |   2012 .081 |   2000   2002   2004   2006  
> 2008   2010     8 |   3 .090 |   2484 .086 |   2460   2484   2524
>   9 |   3 .087 |   3110 .079 |   2914   3108   3110
>  10 |   1      |   2332      |   2332
>
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> 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
>


-- 
------------------------------------------------------
Ran Friedman
Postdoctoral Fellow
Computational Structural Biology Group (A. Caflisch)
Department of Biochemistry
University of Zurich
Winterthurerstrasse 190
CH-8057 Zurich, Switzerland
Tel. +41-44-6355593
Email: r.friedman at bioc.unizh.ch
Skype: ran.friedman
------------------------------------------------------