VMD-L Mailing List

From: Josh Vermaas (vermaas2_at_illinois.edu)
Date: Tue Mar 26 2013 - 15:16:46 CDT

Hi Yarrow,

The fitting routines are trying to find a translation and rotation to
minimize the RMSD, so actually the problems are related, and are caused
by how RMSD is calculated. The formula sums over each atom's deviation
from the reference, and if the number of atoms in the selections isn't
identical, you run into problems, as there isn't a single reference to
compare to. You have two options:

-Concoct a selection that does share the number of atoms (if the number
of protein residues is the same "protein and backbone" should work,
regardless of a changed sidechains)

-Align your structures using a different metric. Multiseq might be just
what you need
(http://www.ks.uiuc.edu/Training/Tutorials/vmd/tutorial-html/node6.html)

-Josh Vermaas

On 03/26/2013 02:10 PM, Yarrow Madrona wrote:
> Hi,
>
> I know this is a basic question but I am having trouble alining two very
> similar molecules (98% seq homology) based on a heme selection. Each time,
> I get the following complaint,
>
> measure fit: selections must have the same number of atoms
>
> A similar result happens when I try to use the RMSD caclulator extension.
> Any help would be appreciated. Thank you.
>