[gmx-users] g_cluster and loops dynamics

[James Starlight]

Dear Gromacs Users!

I'd like to cluster ensemble of the conformations of the long (30 aa) loop
using g_cluster method and one of its clustering methods based on the RMSD
cut-off. Briefly during the md run I've observed different secondary
structure elements formations within the loop- so I'd like to cluster this
segment based on the RMSD and SS criteriumd.

1) From the tutorial I've seen that gromos clustering method is commonly
used. I wounder to know  what another methods could be also specified
regarding my specified (long loop) case because gromos clustering didn't
produced good distribution of the conformers along the several clusters:
using 0.3 cutoff I've obtained ~ 100 clusters from which there were no
significant difference in loop conformations, using big cutoff (>2 A) only
1 cluster have been determoned from the whole ensemble.

2) What reasonable information could provide me transitions between
clusters diagrams ? ( the default xpm file are not vell readable ).

Thanks for help,

James