[gmx-users] CLUSTERING:Number of clusters as function of time.

[Mr.Sridhar]

Hello,

I wanted to know the number of conformations the protein has taken in my
MD run trajectory. For this I am clustering the snapshots using g_cluster.
I mainly wanted to know how the number of clusters varies with time.

In the result file (clust-id.xvg), I found the number of clusters increase
and decrease. To my knowledge, the number of clusters should steadily
increase with time and then reach a plateau where the system visits no
more new conformations.

Could any body explain why the number of clusters
show fluctuations of high and low values instead of gradually increasing
with time?

Regards,

sridhar