On Mon, May 4, 2015 at 10:35 AM, newamber list <newamberlist.gmail.com> wrote:
> Another way someone suggested me is Clustering analysis. As far as I
> googled this term, one gets few clusters and one can choose lowest RMSD
> structure from centroid of most populated cluster. I am just wondering
> would this be a good way of choosing best representing structure?
Representative structures (which have lowest cumulative distance to
all frames in a cluster) are already output by the
'repout'/'singlerepout' keywords. Centroids are not written, but you
can think of average structures as the centroid where coordinate
metrics are concerned.
> Also cpptraj manual suggests there are two algorithms to choose. Which one
> I should use in my case. Also, when someone says 'distance from centroid'
> does it usually means RMSD? I am confused with terminologies used.
There are actually 3 clustering algorithms in cpptraj now: kmeans,
hieragglo, and dbscan. In the context of clustering, the distance
metric is RMSD by default - if you choose something else (like 'data')
then the distance refers to that.
-Dan
>
> Thanks for any help.
> Jiom
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
--
-------------------------
Daniel R. Roe, PhD
Department of Medicinal Chemistry
University of Utah
30 South 2000 East, Room 307
Salt Lake City, UT 84112-5820
http://home.chpc.utah.edu/~cheatham/
(801) 587-9652
(801) 585-6208 (Fax)
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon May 11 2015 - 13:30:03 PDT
