issue for the failing nets for single metal

[kjappelbaum]

i think the issue is that the branching sites are not in the node molecule, this might make an issues if we then do the matching

[kjappelbaum]

the issue might be that the shortest path between metal and the branching site does not include the "complete" Hs and so on - perhaps we should switch from shortest path to something like BFS or all paths

[kjappelbaum]

perhaps all_simple_paths can do the job

[kjappelbaum]

all_simple_paths didn't do the issue, but neither did adding the neighbors.

I'm not sure anymore if this is the problem

[kjappelbaum]

rsm3434 is PIQMUZ_clean

[kjappelbaum]

the extra C i see does not appear in the crystal structure and seems to be a bug

[kjappelbaum]

3434 is indeed a diamondid network as the linker is an edge

[kjappelbaum]

perhaps our simplification doesn't work correctly here

[kjappelbaum]

ok, the results show that the issue might be that we do not have all the relevant branching coordinates in this node

[kjappelbaum]

indeed, we only see two, which is wrong

[kjappelbaum]

but we have 4 original graph branching indices

[kjappelbaum]

this was the issue, adding the branching site back fixed it

[kjappelbaum]

rsm3434 is relatively sensitive to symmetrization
i do not like that