Consider the introduction of aggregates

[ppravatto]

In the past, we discussed the idea of defining aggregates of molecules. An aggregate would be a sort of generalized System object composed of a list of inner System objects. As examples, consider a water cluster or a protein with a quaternary structure: the aggregate is composed of more than one molecular System interacting with non-covalent bonds.

We should discuss:

1. Is there a need of implementing aggregates? Are they advantageous in the definition of computational workflows?
2. Can the aggregate be subject to computation that returns properties that can be partitioned between aggregates?