I've got multiple structures to maintain in my application. All link to the same records, and one of them could be considered the "master" in that it reflects actual relationships held in files on disk. The other structures are used to "call out" elements of the main design for purchase and work orders. I'm struggling to come up with a pattern that deals appropriately with changes to the master data.

As an example, the following trees might refer to the same data:

A
|_
  B
  |_ 
    C
    |_
      D
|_
   E
   |_
     B
     |_
       C
       |_
         D


A
|_
   B
   E
C
|_
  D

A
|_
  B
  C
  D
  E

These secondary structures follow internal rules, but their overall structure is usually user-determined. In all cases (including the master), any element can be used in multiple locations and in multiple trees. When I add a child to any element in the tree, I want to either automatically build the secondary structure for each instance of the "master" element or at least advertise the situation to the user and allow them to manually generate the data required for the secondary trees.

Is there any pattern which might apply to this situation? I've been treating it as a view problem, but it turns out to be more complicated than that when you look at the initial generation of the data.