[Chimera-users] needing help for separating structures

Elaine Meng meng at cgl.ucsf.edu
Mon Nov 26 11:28:34 PST 2018

Hi Yu Guo,
Well, it all depends on how these two states are described in the file.  I took a look at the associated paper in Nature and they don’t really say exactly how they described the two states, but they do mention modeling them at reduced occupancy.  So my best guess if one structure has more than one state, they are described in the file is with “alternate locations” for some atoms (i.e. multiple positions for a given atom) with reduced occupancy values.  For me, it is harder to figure out with mmCIF file than PDB because I’m more familiar with looking at PDB files in a text editor.

Usually when there are alternate locations, they are A and B.  So, you could select (or “delete” instead of “select” but then there is no going back other than by re-opening the original file) all atoms with alternate locations A with command:

select @.A

… and similarly for alternate location B.  In my tests, it looks like A has the lower occupancy than B, and selecting alternate location A or B will each select 2188 atoms. You could also select by occupancy value; however, I get different numbers depending on the value I choose, presumably omitting some atoms with alternate locations or including some of the atoms that do NOT have alternate locations, so be careful using that method:

select @/occupancy<0.4
… I avoid using “=“ because rounding issues may prevent exact match
select @/occupancy<0.45
… or you can use a range, e.g.
select @/occupancy>0.5 and occupancy<0.75

Because they did not clearly state in the paper (or at least I didn’t see it), I could not be sure, but guessing that alternate location A is the minority state and alternate location B is the majority state, and atoms without alternate locations were the same in both states.

How I figured out that alternate location A appears to have lower occupancy than B is with these commands:
disp :ala
labelopt info occupancy
label :ala at .A
label :ala at .B

I hope this helps,
Elaine C. Meng, Ph.D.
UCSF Chimera(X) team
Department of Pharmaceutical Chemistry
University of California, San Francisco

P.S. ChimeraX is much faster reading mmCIF than Chimera.  Also it automatically only uses the alternate location set with average higher occupancy (in this case B).  However, if you want to show the altloc set with lower occupancy, there is no easy/convenient method for it yet, although it can be done with a setattr command.

> On Nov 24, 2018, at 11:28 AM, Yu Guo <yg2 at kth.se> wrote:
> Dear Sir/Madam,
>          I am using Chimara to display large proteins. It is very nice. 6DHO(PDB structure, see attachment) is a very recent PSII structure consisting both S2 and S3(richer) states, which are highly overlapped in the raw data. I am urgently wondering how can I separate S3 out from the S2/S3 mixture by Chimera, especially the OEY​ cluster and all its ligands? Thank you very much!
> Best wishes
> <6dho.cif>

More information about the Chimera-users mailing list