Hi Elaine,

Thank you very much for your answer, it is very helpful.!😁
 I have another doubt:
I use Chimera to predict structures using alphafold. When the session is completed, I get a folder with the different PDBs. Usually these PDBs are named as: af1024_unrelaxed_rank_001_alphafold2_ptm_model_4_seed_000.pdb , I get 5 different PDBs whose rank is between 001 and 005. Then I have a best model.pdb, 
Is this best model the rank_001.pdb? 
Would I be doing something wrong if I choose a different rank that matches more closely with lab data where I can calculate whether a structure is open or closed? I mean, I make the prediction, the best model gives me a "closed" structure and my lab data indicates an "open" structure. But a pdb with rank_002 or _003 shows open structure, is it possible to choose that model before than best model?

I don't know if I have explained myself, my English is not very good 😅. Thank you very much for your time and help. 

Best 

Diego

Dr. Diego Angosto Bazarra
Molecular inflammation and experimental surgery group. 
IMIB. University Hospital Virgen de la Arrixaca



El 2 jul 2024, a las 18:17, Elaine Meng <meng@cgl.ucsf.edu> escribió:

Hi Diego,
As far as I know there is no measurement (in any program, I mean) that is the % similarity in structures.  There are lots of other structure similarity measures.  What people use the most is probably RMSD and that is already reported in the Log after you use matchmaker.   Matchmaker uses only one atom per residue, CA atoms in amino acids, to calculate this RMSD. 

For example, these commands:

open 2gbp
open 2fw0
mm #2 to #1 show true

... open two structures and run Matchmaker, and result is shown in this screenshot.  The "show true" shows the sequence alignment and the positions used for the final fit are outlined in orange boxes on the alignment.  The Log window shows that the RMSD for the 147 positions (CA-CA pairs) in the orange boxes is 0.425 angstroms, and also across the entire sequences (305 CA-CA) pairs is 7.702 angstroms.  The gray histogram over the sequences shows the CA-CA distance at each position.

<Screen Shot 2024-07-02 at 9.14.41 AM.png>


If you want to use more atoms (not just one CA atom per residue) to calculate the RMSD without moving the structures from their current positions, then you could use the "rmsd" command.  See:

Elaine
-----
Elaine C. Meng, Ph.D.                       
UCSF Chimera(X) team
Resource for Biocomputing, Visualization, and Informatics
Department of Pharmaceutical Chemistry
University of California, San Francisco


On Jul 2, 2024, at 2:12 AM, DIEGO ANGOSTO BAZARRA via ChimeraX-users <chimerax-users@cgl.ucsf.edu> wrote:

Hello,

My name is Diego. I’m a beginner user of ChimeraX (1.8) and I was wondering if there is an option to check the % of similarity between two structures after  making the matchmaker.

I mean, I make a matchmaker with two structures, can I see the % of similarity between these two structures? Not between the sequences, but between the structures.

Thank you very much for your time and attention.

Best regards

Dr. Diego Angosto Bazarra
Molecular inflammation and experimental surgery group. 
IMIB. University Hospital Virgen de la Arrixaca