findclash misleading and wrong

[Elaine Meng]

ChimeraX findclash gives different results from Chimera findclash and it is not easy for a user to figure out why. I understand that certain things may not be implemented yet, but it should be clearer what's different. For example, the following gives 19 contacts in Chimera and 108 contacts in ChimeraX.

open 1zik
findclash protein overlap 0 hb 0

(Difference #1) ChimeraX uses “test self” as the default and does not let the user enter a “test” option (Chimera default is “test other”). Log should report which "test" option is being used. Ideally "test other" would be implemented (I could be wrong but guess it wouldn't be much work) and made the default. Further,

ChimeraX: findclash protein ov 0 ... gives 11 contacts
Chimera equivalent: findclash protein ov 0 test self ... gives 56 contacts

(Difference #2) ChimeraX apparently subtracts "hb" value from the overlap of every pair of atoms even if they are not potentially H-bonding. Default "hb" value is 0.4, as reported in the log (or if user specifies a value, that value is used and reported in the log). Instead should say "hb" is not implemented and set it to 0 and report it as such. (Last time I tested it was wrong in a different way: it claimed the value was 0.4 or whatever the user specified, yet always really used 0.)

[Conrad Huang]

Fixed in 094e643.

"other" is still not implemented, but using it causes a LimitationError.

Non-zero hbond_allowance causes a warning message to be displayed and is set to 0.

[Elaine Meng]

Don’t know how much of a fix this is, since using the command gives a traceback now.  The “test other” option is the default, but it gives the traceback.  Thus one has to explicitly use a different “test” value.  Unfortunately “test atomspec” and “test model” are not implemented either and give a similar traceback. The only allowed value is “test self” … so including the lack of hbond correction, it isn’t particularly usable.  I haven’t documented it.

You said:
“ other is still not implemented, but using it causes a LimitationError” 
… so does that mean the traceback is on purpose??

Example:
open 1zik
findclash protein
Traceback (most recent call last):
File "/Users/chimera/Applications/ChimeraX_Daily.app/Contents/Library/Frameworks/Python.framework/Versions/3.5/lib/python3.5/site-packages/chimerax/cmd_line/tool.py", line 158, in execute
cmd.run(cmd_text)
File "/Users/chimera/Applications/ChimeraX_Daily.app/Contents/Library/Frameworks/Python.framework/Versions/3.5/lib/python3.5/site-packages/chimerax/core/commands/cli.py", line 2255, in run
result = ci.function(session, **kw_args)
File "/Users/chimera/Applications/ChimeraX_Daily.app/Contents/Library/Frameworks/Python.framework/Versions/3.5/lib/python3.5/site-packages/chimerax/exp_cmd/findclash.py", line 29, in findclash
raise LimitationError("findclash test \"%s\" not implemented" % test)
chimerax.core.errors.LimitationError: findclash test "other" not implemented

[Elaine Meng]

OK I opened a new ticket #518