When I look at the ChimeraX Segger code in the segment_dialog.py file in the Overlap function https://github.com/tomgoddard/segger/blob/2043b3279feb070632146debd7d0fd8496... I see that to select a region it looks like it requires at least 50% of the grid points within that region are occupied by selected atoms. Maybe in your case the selected atoms occupy only a small fraction of the segmentation regions you want to select. You could change the threshold, ovRatio in the code (currently 0.5) to something smaller like 0.01 and maybe it would give you all regions that overlap even a few of your selected atoms. That segment_dialog.py Python file is in your ChimeraX distribution and you can just edit it with a text editor and restart ChimeraX. On a Mac that file would be found in ChimeraX.app/Contents/lib/python3.9/site-packages/chimerax/segger/segment_dialog.py On Windows and Linux you would need to find the ChimeraX lib directory and the path would be similar. The Segger segmentation tool in ChimeraX and Chimera was developed by Greg Pintilie. He may have a better answer. Tom

On Jun 15, 2023, at 4:24 PM, kizziah4--- via ChimeraX-users <chimerax-users@cgl.ucsf.edu> wrote:

Is there some degree to which a segment must overlap an atomic model to be selected with "Over Sel. Atoms"? Can it be changed?

I frequently use Segment Map to create custom masks for focused refinement and classification and for signal subtraction. I tend to use reference atomic models to expedite this process, but I am having problems selecting segments overlapping atomic models (compared to older versions of Chimera). I am on Windows 10 with ChimeraX version 1.5 (2022-11-24). I have experienced the same on Ubuntu-20.04 with ChimeraX v1.6.1.

For example, I have an unsharpened map estimated at 4.8A resolution from RELION and an atomic model of the same structure downloadeded (via ChimeraX) from the PDB. After deleting some unneeded chains and duplicating others, fitting the model into the map using simulated density (4.8A resolution, 0.77 correlation), segmenting the map at a low threshold that covers all (but a couple of loops) of the atomic model, ungrouping the segments once for granularity, selecting the entire atomic model, and clicking "Over Sel. Atoms", no segments are selected.

If I try repeating the process with the backbone only (0.80 correlation), no segments are selected. I have succeeded in some cases by ungrouping all segments ~4 times, but in this particular case it only selects a few segments corresponding to parts of some alpha helices. _______________________________________________ ChimeraX-users mailing list -- chimerax-users@cgl.ucsf.edu To unsubscribe send an email to chimerax-users-leave@cgl.ucsf.edu Manage subscription: