[gmx-users] rmsd b/w average structure and the trajectory
feenstra at chem.vu.nl
Thu Oct 16 09:01:01 CEST 2003
> Dear All,
> I had a 4ns md and the simulation seems to be stable. I calculated the
> average structure using g_rmsf -ox and then looked at the rmsd between
> the averaged structrure and the frames that were used to calculate the
> average. The rmsd of the backbone is all at around 2nm. This is kind of
> way off after the average. I was trying to get a representative
> structure for the md. Now it seems I have to pick a frame that has the
> smallest rmsd to the average structure but at around 2nm range. I
> certainly do not feel very comfortable with that. Any suggestions?
> P.S. If the avereage works so bad, maybe we should bot bother using it.
> I have seen some disscusions on this issue before. Some suggested that
> energy minization of the average could generate a representative
> structure. How about just picking a frame when the simulation is stable
> and minimize that? Just a thought. Any input is appreciated.
IMHO, if you are looking for a representative structure, you should be looking
at clusters. g_cluster can do that for you, it will calculate RMSD's, or accept
a matrix (.xpm format, from e.g. g_rms), and cluster low-rms pairs together.
Choosing a certain rmsd cut-off, and looking at the (middle structure of the)
largest cluster(s?) should give you insigight into what might be a 'representative'
cluster and/or structure in your simulation.
| | |
| _ _ ___,| K. Anton Feenstra |
| / \ / \'| | | Dept. of Pharmacochem. - Vrije Universiteit Amsterdam |
|( | )| | | De Boelelaan 1083 - 1081 HV Amsterdam - Netherlands |
| \_/ \_/ | | | Tel: +31 20 44 47608 - Fax: +31 20 44 47610 |
| | Feenstra at chem.vu.nl - www.chem.vu.nl/~feenstra/ |
| | "If You See Me Getting High, Knock Me Down" |
| | (Red Hot Chili Peppers) |
More information about the gromacs.org_gmx-users