VMD-L Mailing List

From: Visvaldas Kairys (kairys_at_uma.pt)
Date: Fri Mar 24 2006 - 03:45:51 CST

Thank you,

I think duplication of the atoms will be a satisfactory workaround for me!
Actually, best is to duplicate ALL the atoms instead of just one
because the RMSDs reported will be _correct_ (as if the atoms were not
duplicated, but the without the "bug")! -
I tried to experiment with
another molecule containing 9 atoms, and the RMSD was identical for 9 vs 9
alignment and 18(duplicated) vs 18(duplicated) alignment, which is not
surprising due to the definition of RMSD.

Vis

On Thursday 23 March 2006 23:22, you wrote:
> Hello,
>
> I replicated this problem as well using 1.8.3. However, if I remember
> correctly the algorithm itself requires a minimum of 4 atoms to match.
>
> If you just duplicate one atom each for the two molecules the alignment
> works just fine.
>
> Brian
>