Check use of WaterBcutoff parameter in Reduce

[russell-taylor]

This is set to 40 and used to cull waters along with occupancy in line 1679 of reduce.cpp. Has this check made it into the new code anywhere?

The corresponding probe2.py code has values of 0.25 and 80 whereas reduce.cpp has 0.66 and 40.

[russell-taylor]

These are not currently being handled. The corresponding ignoring of atoms is happening in the probe2 app and not in the Probe2 library.

(nope) Add Phil parameters for the occupancy and B-factor cut-off and put the filtering in after line 290 in Optimizers.py; remove offending atoms from the list. Actually, the command-line options to change these values are not used in MolProbity, so we don't need to expose them as parameters though we do need to implement them so that we get the same optimization behavior. Added @todo comments to add these as parameters later.

[russell-taylor]

(done) Actually, the water Oxygens are ignored in addition to the hydrogens on the Oxygens. This should be done in Reduce2 but not Probe2. This can be done inside the loop where we have identified water Oxygens before calling getPhantomHydroensFor() on it. Instead, we can mark the atom as not to be used by removing them from the spatial query structure and from the list of atoms to be considered.

[russell-taylor]

@todo: It may be a bug that Probe did not remove the Oxygen atoms the same way that Reduce did. If this is the case, we may want to replace the minimum_polar_hydrogen_occupancy and maximum_polar_hydrogen_b Phil parameters in Probe2 with water instead of polar_hydrogen and have it remove the Oxygens from consideration when doing scoring. This will get rid of some of the clashes found in Fe_1brf_rubredoxin_0.95A_reduced.pdb > C:\tmp\Fe_1brf_rubredoxin_0.95A_reduced.pdb because the Water Oxygens will be ignored due to too-low occupancy.