[chimera-dev] atom order in RMSD calculations

[Elaine Meng]

Hi Kristen,
I'm CCing chimera-users since this is more of a user question than developer/programmer (chimera-dev) issue.

The serial number is not the main factor in pairing the atoms.  As mentioned in the "rmsd" documentation,
<http://plato.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/rmsd.html>
"If atom order is not specified...
the atoms within a residue are ordered first by name, and where these are identical, by alternate location identifier, and where these are also identical, by serial number." 

As for residues, if your models have multiple residues I believe they are compared in order.

However, if your structures have exactly the same atom names and in the same order, I believe the rules above would pair the atoms in the epected order.

The only other issue I can think of is that chemically symmetrical pairings are not tried, for example, if you had two phenyl rings or two carboxylates that exactly match in space but the atom names are in "flipped" positions, you would still get a higher RMSD.

If it seems Chimera is not pairing the atoms as advertised, please use "Help... Report a Bug" in the Chimera menu and attach a session file with the two models, and give a short description and an email address if you'd like to hear back.

There are a couple of ways to explicitly give the atom order, but if your structures have many atoms, they would be difficult/tedious:

(a) interactively pick the atoms one by one from the graphics display in the correct order and then use the word "sel" to specify them in the command, as described in more detail in the documentation (link above)

(b) specify all the atom names in order in the command

I hope this helps,
Elaine
-----
Elaine C. Meng, Ph.D.                       
UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab
Department of Pharmaceutical Chemistry
University of California, San Francisco



On Sep 10, 2011, at 11:44 AM, Kristen Eisenberg wrote:

> [chimera-dev] Calculating RMSD
> 
> Hi,
> I want to calculate the RMSD between two mol2 structures in place. The two
> structures are different forms of the same structure. I haver modified the
> mol2 so that the two of them have the same serialNumber in their respective
> atoms. The problem is that when I use the command rmsd #model1 #model2, it
> return an incorrect value of the rmsd, as the value showed is too high,
> about 5, when it should be about 1 (the two structures are very similar). Is
> there any way to calculate the RMSD value in place correctly or I am doing
> anything wrong?
> 
> Thanks for your help.
> Kristen Eisenberg