[gmx-users] rmsd-dist.xvg

[X.Periole]

On Mon, 13 Feb 2006 10:55:18 +0200
  Itamar Kass <ikass at cc.huji.ac.il> wrote:
> 
> Thanks Xavier,
> 
> If I get a two picks graph, is this mean I have two 
>distinct clusters? (for an
> arbitrary RMS)?

That would be nice if it was that easy !!! You probably 
have
one pick at low rmsd values ... and another one at higher 
values, right. That means only that you have two different
populations of the rmsd values.

Within the first population is included all the values
of rmsd that will help you building clusters basically all
the pairs of structures good candidates to be part of one
of the clusters. The second pick are all the pairs that 
will
not help building the clusters. That is onw wayt to see 
it.

Note that two structures can be at 2 ang from a reference
but be at 4 ang from eachother.

The point is that the minimum between the first and second
is generally used as cutoff to build the clusters.

To get the number of clusters you have to run the program
again giving a cutoff, the minimum between the picks or
another one.

XAvier