[gmx-users] g_rms on large amount of data

[khuchtumur bumerdene]

Hi,
I'm using Gromacs 4.5.5 to simulate a protein in explicit solvent.
The simulation is running fine and I have collected 9 100ns simulations.
I'm eliminating the first 50 ns for equilibration and sheer data size.
What I want to do now is to cluster the simulations based on rmsd of
certain residues. I have used make_ndx to select the residues and made a
group.
I have used trjcat -cat option to concatenate the 9 runs into a single
giant .xtc file. In the .xtc file, I only output the protein.
My plan was to use g_rms to create a rmsd.xpm table and using that with
g_cluster. However, when I do run g_rms, it seems like it will take weeks
to finish.
The RMSD matrix size becomes 225000x225000 and my question is: How do you
cluster this much data? Any recommendation would be useful.

Thank you for your time.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20111213/e1888aeb/attachment.html>