User blog comment:Wythagoras/Extension to Friedmann's stuff/@comment-1605058-20130612190549

I don't think there is any agreed way to define subsequence in many dimensions. For example, is (2,2;2,2) subsequence of (2,2,1;1,1,1;1,2,2)? Depending on definition, n(a,b,c) may or may be not defined.

There are no proven results extending Robertson-Seymour theorem to labelled graphs. Label of edge contracted to vertex may be ambiguous. If we limit ourselves to topological minors, there is still no proof that this relation is well-quasi-order.