Overview of Established Fragment Based Methods

TODO: Proper overview instead of bullet points

• Many fragment-based methods started in relative isolation

• Each group called their method something different

• Even when similarities emerge, typically the gropu maintains the method’s name

• Tendency to talk about methods as if they’re black box methods

  • e.g., much in the way that one would say “we ran B3LYP” groups will say they “ran <insert fragmentation method here>” with little to few details.

• Unlike established quantum chemistry methods, fragment based methods tend to be much more sensitive to the specifics.

• Our argument is that each method can be decomposed into a few key algorithms

• Specifying the algorithms is akin to specifying a method.

• Upon enumating the algorithms we can propose new method names which are more in the spirit of black box methods (although the utility of such names over just specifying the algorithm pieces is debatable).