[Chimera-users] Question about RMSD calculation of MatchMaker

[Elaine Meng]

Hi Matthias!
As shown in your image of the dialog, you’re using the “Iterate by pruning…” option that removes farther-apart pairs from the fit.  That’s why you could get different numbers of pairs used to calculate the RMSD even if all chains contain the same numbers of residues.  Iteration is described in the MatchMaker docs:
<http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/matchmaker/matchmaker.html#iterate>

If you turn off the “iterate” option, it will not try to improve the fit by pruning and will instead use all the pairs (CA-CA atom pairs of residues aligned in the sequence alignment).  If the structures include the same residues for each chain, the pairwise chain-chain RMSDs will all be calculated from the same number of CA-CA atom pairs.

The dialog image also shows you are fitting A-B, A-C, A-D.   For completeness you may also want to try B-C, B-D, C-D.

Another issue is that MatchMaker or its command equivalents (mmaker, matchmaker) only use one atom per residue, CA, and only considers aligned biopolymer chains (peptides, nucleic acids).  If that’s what you want, no worries.  However, you could use the “match” command instead and specify any sets of atoms for RMSD calculations, such as all backbone, with or without iteration.  However, it’s harder to use since you have to specify the atoms in the command line yourself, unlike MatchMaker that figures out which atoms to pair automatically.  You could include waters, but it would be tedious/difficult to specify each water residue by name, in the corresponding orders, for the two structures to be matched.  Also the waters would have to be in the respective models. 

Matchmaker vs. match is discussed in the following link:
<http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/superposition.html>

“match” command manpage, examples:
<http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/match.html>

command-line atom specification:
<http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/atom_spec.html#basic>

another “match” example in a tutorial, see the Matching section in:
<http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/tutorials/frameimages.html>

If you are interested in the variation over the length of the chain as opposed to getting a single number, a nice way to show that is to first superimpose the chains, then associate them all with a single copy of the sequence.  Then over the sequence you can display an RMSD histogram showing the alpha-carbon (or other atoms, depending on which RMSD you choose) structural variability at each position in the sequence.  To do that:

(1) superimpose chains however you like (Matchmaker or match)
(2) show sequence of any one of them, presumably they are basically the same (menu: Favorites… Sequence, choose one)
(3) in the sequence window, use Structure… Associations to associate all copies of the structure with the single sequence
(4) in the sequence window, use the Headers menu to show the RMSD of interest (and hide other headers you don’t want to see)

You can even show these RMSD values with colors or worm thickness on one of the copies, as described in more detail here:
<http://plato.cgl.ucsf.edu/pipermail/chimera-users/2014-March/009710.html>

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 Oct 1, 2014, at 9:38 PM, Matthias Fellner <mfellner at chemistry.otago.ac.nz> wrote:

> Good day
>  
> I am a PhD student at the University of Otago New Zealand. First I would like to thank you for this very useful software, I use it for all my protein crystallography pictures and a lot of analysis as well, I will use a citation for sure when I publish using your software.
> A current analysis of mine involves a crystal structure that has 4 chains in the asymmetric unit. They chains are very similar when aligned and I would like to have a statistic to show this.
>  
> I assume the RMSD is the way to go. So I saved each chain as its own pdb, loaded them in Chimera and used your MatchMaker to align chain B,C and D to chain A (randomly chosen because of notation of chains).
>  
> The chains have an identical amino acid sequence (~200 residues), I just build 1-5 residues more or less at each end of the chain but otherwise they should all be the same. Now the reply log gives out the RMSD and the number is under 1Å as I expected and I can quote that. But my question would be why did it use a different numbers of atom pairs for the different chains (191 or 188 or 193) and which atom pairs are those. I assume it picks one atom from the main chain of each residue and the different number comes from the difference residue number at the end of the chains.
>  
> I searched through your help files and with google through your board but could not find the answer to what atoms are paired and if you can change the pair for example to include all atoms of the mainchain or even all atoms all together.
>  
> Sorry for the long question but I wanted to give all the details, attached also the (default) options I had for the MatchMaker and the reply log.
>  
> In addition a small question which I guess wont be possible – is there any way to also align water molecules adjacent to the chains although all water molecules are saved in chain S together?
>  
> Thank you
> Matthias
>  
>  
> <image001.png>
>  
> Matchmaker chainA.pdb, chain A (#0) with chainD.pdb, chain D (#3), sequence alignment score = 999.7
> with these parameters:
>                 chain pairing: bb
>                 Needleman-Wunsch using BLOSUM-62
>                 ss fraction: 0.3
>                 gap open (HH/SS/other) 18/18/6, extend 1
>                 ss matrix:  (O, S): -6 (H, O): -6 (H, H): 6 (S, S): 6 (H, S): -9 (O, O): 4
>                 iteration cutoff: 2
> RMSD between 191 atom pairs is 0.506 angstroms
>  
>  
> Matchmaker chainA.pdb, chain A (#0) with chainC.pdb, chain C (#2), sequence alignment score = 999.1
> with these parameters:
>                 chain pairing: bb
>                 Needleman-Wunsch using BLOSUM-62
>                 ss fraction: 0.3
>                 gap open (HH/SS/other) 18/18/6, extend 1
>                 ss matrix:  (O, S): -6 (H, O): -6 (H, H): 6 (S, S): 6 (H, S): -9 (O, O): 4
>                 iteration cutoff: 2
> RMSD between 188 atom pairs is 0.551 angstroms
>  
>  
> Matchmaker chainA.pdb, chain A (#0) with chainB.pdb, chain B (#1), sequence alignment score = 987.2
> with these parameters:
>                 chain pairing: bb
>                 Needleman-Wunsch using BLOSUM-62
>                 ss fraction: 0.3
>                 gap open (HH/SS/other) 18/18/6, extend 1
>                 ss matrix:  (O, S): -6 (H, O): -6 (H, H): 6 (S, S): 6 (H, S): -9 (O, O): 4
>                 iteration cutoff: 2
> RMSD between 193 atom pairs is 0.336 angstroms
> _______________________________________________
> Chimera-users mailing list
> Chimera-users at cgl.ucsf.edu
> http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users