[Chimera-users] RMSD Chimera

[Elaine Meng]

Hi Ishita,
All of these questions are already answered in the Help and/or clear from the what the Reply Log says, but I will try to summarize here.  Next time please just click the Help button on the tool and read the information first, before asking.

(1) <http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/matchmaker/matchmaker.html#matching>
"Fitting uses one point per residue: CA atoms in amino acids and C4' atoms in nucleic acids."

(2) after you run matchmaker, the Reply Log (open from Favorites menu) gives a message like

"RMSD between 147 pruned atom pairs is 0.425 angstroms; (across all 305 pairs: 7.702)”

… so as it already says, the first value is for the pruned set, the second is all the positions aligned in the sequence alignment, which you can display with the matchmaker option to show the pairwise alignment.

(3) The main purpose of the Match->Align tools is to create a multiple sequence alignment.  You could only get pairwise sequence alignments from Matchmaker, but if you superimposed 3 or more proteins, then you can use Match->Align to get a single sequence alignment of those 3 or more sequences.  However, it also reports RMSD values as additional information after it makes the sequence alignment.  The Reply Log will have something like

Match→Align cutoff: 5.0, in column if within cutoff of: any
207 residue pairs aligned
207 fully populated columns

Evaluating superpositions across all 207 fully populated columns in the final alignment:
RMSD of 2gbp, chain A with 2fw0, chain A: 1.704
Overall RMSD: 1.704
Sequence lengths: 309 305
SDM (cutoff 5.0): 40.695
Q-score: 0.344

… in that case the RMSD is over the number of “fully populated columns” meaning all the positions where all sequences have a residue (not a gap) in the output multiple sequence alignment, in this case 207 positions. 

<http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/matchalign/matchalign.html#measures>
"Match -> Align reports root-mean-square deviations (RMSDs) calculated using one atom per residue: CA in amino acids, C4' in nucleic acids.”

However, if you only have 2 proteins and their sequences are similar enough that they are easy to align directly by Matchmaker, there is not really any reason to use Match->Align. 

(4) Multalign Viewer does the same thing with Structure… Match, it just uses the CA atoms of the residues that are in the same columns of the sequence alignment.
<http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/multalignviewer/multalignviewer.html#superposition>
"Structure... Match in the Multalign Viewer menu allows structures to be superimposed based on the alignment of their associated sequences. […] Fitting uses one point per residue: CA in amino acid residues and C4' in nucleic acid residues.”

Best,
Elaine
-----
Elaine C. Meng, Ph.D.
UCSF Chimera(X) team
Department of Pharmaceutical Chemistry
University of California, San Francisco

> On Jul 21, 2019, at 9:41 AM, Ishita Gupta <ishita at icgeb.res.in> wrote:
> 
> Hi,
> I have a few questions in mind regarding rmsd calculation in chimera:
> 
> 1. Does the matchmaker tool calculate c alpha?
> 2. Do we consider RMSD of pruned atoms or the total atoms while reporting?
> 3. If we use structural comparison match align tool then rmsd is based on multiple sequence alignment? And it also gives C alpha?
> 4. If we calculate rmsd from multi align viewer by clicking on the structure and then match. What these rmsd values denote?
> 
> Please elaborate on these points. 
> Thanks
> Ishita