[Chimera-users] color by RMSD
meng at cgl.ucsf.edu
Wed Apr 4 10:17:13 PDT 2012
Absolutely you could, but it is not as simple... There is an "rmsd" command that measures RMSD of whatever atom sets you specify, in their current positions (without fitting). So after you superimposed the structures using chain B only (Matchmaker GUI and mmaker command both have options to use specific chains for superposition), then you would need to run a series of commands something like:
rmsd #0:1.a #1:1.a
... calculate RMSD between all atoms of model 0 residue 1 in chain A vs. model 1 residue 1 in chain A
The two residues would have to be the same type (contain the same number of atoms). If the numbering is different in the two structures, you'd have to adjust for that, e.g.
rmsd #0:1.a #1:9.a
Then you'd have to write all the results from the Reply Log and process them into the appropriate attribute definition file format, and then read that file in to assign the RMSD values as attributes, as described for method (B) in my previous response.
Actually there is one new spiffy way to loop through a residue range with commands (Chimera 1.7, so daily builds), for example if you wanted to measure RMSD of model 0 chain A residues 1-85 vs. model 1 chain A residues 6-90:
perframe "rmsd #0:$1.a #1:$2.a" range 1,85 range 6,90
If there are gaps in the sequence alignment of the two proteins, you'd need a separate perframe command for each contiguous nongapped segment of alignment, since their numbering offset would change. Also if there are partial residues (due to missing density) or residues with alternate locations, there will be errors since the numbers of atoms in the two residues wouldn't be the same. You'd have to adjust your script to skip missing or partial residues or change to using only backbone atoms if the sidechain atoms are missing. I'd check for and delete alternate locations beforehand (for example, the Dock Prep tool has an option to do that, or you could delete all alternate locations B with "delete @.b"). "perframe" instructions:
For example, I can superimpose 2gbp and 2fw0, remove alternate locations B, and then calculate the all-atom RMSDs for residues 3-300 vs 3-300. I would only have to use one substitution variable and range since these two structures have the same numbering:
mm #0 #1
perframe "rmsd #0:$1.a #1:$1.a" range 3,300
The reformatting of results into attribute assignment format would probably be the most tedious step, and possibly could be done with your own scripts. For example, the Reply Log would contain something like this
Executing rmsd ['#0:299.a', '#1:299.a']
RMSD between 8 atom pairs is 6.970 angstroms
... which you would want to reformat to lines like this in the assignment file:
I hope this helps,
Elaine C. Meng, Ph.D.
UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab
Department of Pharmaceutical Chemistry
University of California, San Francisco
On Apr 3, 2012, at 2:36 PM, Thomas Evangelidis wrote:
> Hi Elaine,
> Many thanks for the detailed information! I have managed to use matchmaker and save the RMSD values as you described in method (A), but the problem is that MultialignViewer measures only the RMSD of the CA atoms that were used in the alignment. In my case I want to align chains B of the two structures but measure the RMSD between chains A. Is it possible to do that?
> Thanks in advance for any help.
> On 3 April 2012 19:36, Elaine Meng <meng at cgl.ucsf.edu> wrote:
> Hi Thomas,
> Two general possibilities:
> (A) If the RMSD is to be calculated by Chimera, you would need to superimpose the proteins to calculate the RMSD, then color by that value using Render by Attribute or command rangecolor. However, after the coloring step you could move the proteins apart. The RMSD calculation in Chimera uses CA atoms, so if you only have two proteins, the per-residue "RMSD" boils down to simply the CA-CA distance between the two proteins in the superimposed state.
> (B) Alternatively, if the RMSD is calculated by some other program you could create an "attribute definition" file that lists the values, read the file into Chimera to assign the attribute values to residues (or atoms), and then color by your custom attribute the same way as in (A). The attribute file format is quite simple, see description and examples:
> However, method (A) is probably easiest if the CA-CA distance is an acceptable measure. Example using MatchMaker command (mm) to superimpose and show sequence alignment:
> open 2gbp
> open 2fw0
> mm #0 #1 show true
> ... sequence alignment should appear, including RMSD histogram above the sequences ...
> To define and apply the coloring, you could use Render by Attribute (see the attribute of residues named mavRMSD) or rangecolor, for example:
> rangecolor mavRMSD min purple mid red max yellow
> Then you can un-superimpose the structures, for example, with command: reset, or manually with the mouse.
> I hope this helps,
> Elaine C. Meng, Ph.D.
> UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab
> Department of Pharmaceutical Chemistry
> University of California, San Francisco
> On Apr 3, 2012, at 5:46 AM, Thomas Evangelidis wrote:
> > Dear UCSF Chimera users and developers,
> > Is it possible to color a pair of structures representing different stages of the thermodynamic cycle of a protein by the RMSD, without aligning them?
> > thanks in advance,
> > Thomas
> Thomas Evangelidis
> PhD student
> Biomedical Research Foundation, Academy of Athens
> 4 Soranou Ephessiou , 115 27 Athens, Greece
> email: tevang at bioacademy.gr
> tevang3 at gmail.com
> website: https://sites.google.com/site/thomasevangelidishomepage/
> Chimera-users mailing list
> Chimera-users at cgl.ucsf.edu
More information about the Chimera-users