Open
Description
The current matchAtoms() uses the entire molecule to get MCS, a simple idea to accelerate it is to get MCS for heavy atoms and then match up hydrogens hanging off the MCS.
See this implementation for example:
https://github.com/OpenFreeEnergy/Lomap/blob/84e1bb0d20ceeed835166fe7218dd21f04248030/lomap/mcs.py#L1170
I didn't quantitatively test the speedup(no benchmark found) but it performed very well for large molecules in my project ( the current matchAtoms() often times out).