unique.py

[joegair]

I am getting unexpected behavior with unique.py it characterizes the following four files as not unique using the default RMSD cutoff.

If we set 01_opt as the reference structure, it is outputting RMSDs of

04_opt (RMSD = 0.135)
05_opt (RMSD = 0.022)
06_opt (RMSD = 0.149)

when I open these files in ChimeraX and use the align function align #2 toAtoms #1 I get much larger RMSDs

04_opt (RMSD = 1.887)
05_opt (RMSD = 1.896)
06_opt (RMSD = 2.189)

Maybe I just need to decrease the RMSD tolerance in unique.py. Bringing this to your attention because I did not expect unique.py to describe these structures as not unique and I was surprised by the difference in RMSD assigned by unique.py vs align in chimeraX.

04_no_cat_int_cis_01_opt.txt
04_no_cat_int_cis_04_opt.txt
04_no_cat_int_cis_05_opt.txt
04_no_cat_int_cis_06_opt.txt

[joegair]

Another example of where the results of unique.py are surprising. Using RMSD threshold 0.01 we, unique.py assigns

there are 3 structures similar to 04_opt: grey in screenshot below
07_opt (RMSD = 0.008) red in screenshot below
09_opt (RMSD = 0.000)
13_opt (RMSD = 0.000)
there are 2 structures similar to 05_opt: black in screenshot below

grey-vs-red is assigned identical, despite a 90 degree rotation about the C-C(aryl) bond
grey-vs-black is assigned inequivalent, even though there is only a small nudge to the amide bond.

04_no_cat_int_cis_04_opt.txt
04_no_cat_int_cis_07_opt.txt
04_no_cat_int_cis_05_opt.txt