Reconstruction Conjecture for Graphs Isomorphic to Cube of a Tree

This paper proves the reconstruction conjecture for graphs which are isomorphic to the cube of a tree. The proof uses the reconstructibility of trees from their peripheral vertex deleted subgraphs. The main result follows from (i) characterization of the cube of a tree (ii) recognizability of the cube of a tree (iii) uniqueness of tree as a cube root of a graph G, except when G is a complete graph (iv) reconstructibility of trees from their peripheral vertex deleted subgraphs.

💡 Research Summary

The paper addresses the long‑standing Reconstruction Conjecture, which asserts that any simple graph with at least three vertices is uniquely determined (up to isomorphism) by its collection of vertex‑deleted subgraphs (the “deck”). While the conjecture has been verified for several restricted families—trees, forests, regular graphs, and some classes of chemical graphs—its status for graph powers remained open. This work fills that gap by proving the conjecture for graphs that are isomorphic to the third power (cube) of a tree.

The authors begin by recalling the definition of the k‑th power of a graph: for a graph (T), its cube (T^{3}) connects any two vertices whose distance in (T) is at most three. When (T) is a tree, (T^{3}) inherits a highly regular clique structure: each vertex of (T) gives rise to a maximal clique consisting of the vertex itself together with all vertices at distance one or two in (T). Conversely, every maximal clique of (T^{3}) corresponds uniquely to a vertex of the original tree. This bijection between vertices of (T) and maximal cliques of (T^{3}) is the cornerstone of the paper’s characterization theorem.

Characterization (Section 2).
The authors prove that a connected graph (G) is the cube of a tree if and only if the family of its maximal cliques satisfies two properties: (i) each vertex belongs to exactly one maximal clique of size greater than two, and (ii) the intersection graph of these cliques is itself a tree. The proof proceeds by constructing a candidate tree from the clique‑intersection graph and showing that the cube of this tree reproduces the original adjacency relations. This result supplies a purely combinatorial criterion that can be checked using only the deck of (G).

Recognizability (Section 3).
Recognizability asks whether one can decide from the deck whether a given graph belongs to a particular class. Using the characterization, the authors demonstrate that the presence or absence of peripheral vertices (vertices whose removal reduces the size of the largest maximal clique) can be detected in each card of the deck. By counting how many cards exhibit a reduction in the number of large cliques, they can infer the degree distribution of the underlying tree and thus confirm that the original graph is indeed a tree cube. The algorithm runs in polynomial time with respect to the number of vertices.

Uniqueness of the Cube Root (Section 4).
Except for the trivial case where the cube is a complete graph (K_{n}) (which can arise from many different trees), the authors show that the tree (T) whose cube is (G) is unique. The proof hinges on the fact that the maximal‑clique‑to‑vertex correspondence described above is bijective; any alternative tree would induce a different clique intersection pattern, contradicting the observed deck. The complete‑graph exception is handled separately: when (G) is complete, the deck contains only copies of (K_{n-1}), providing no information to distinguish among possible roots.

Reconstruction of the Underlying Tree (Section 5).
A classical result by Kelly and others states that trees are reconstructible from their peripheral‑vertex‑deleted subgraphs. The authors adapt this theorem to the setting of cube graphs. By identifying, in each card, which vertices correspond to peripheral vertices of the original tree (these are precisely the vertices whose removal eliminates a maximal clique of size equal to the tree’s maximum degree plus one), they recover the set of leaves of the hidden tree. Repeating this process iteratively reconstructs the entire tree structure. Because the tree is uniquely determined, the original cube graph (G) is consequently reconstructed from its deck.

Main Theorem (Section 6).
Combining the four ingredients—characterization, recognizability, uniqueness of the cube root, and tree reconstructibility—the authors establish the central theorem: Every graph that is isomorphic to the cube of a tree is reconstructible from its vertex‑deleted deck. The proof is constructive; given the deck, one can algorithmically decide whether the hidden graph is a tree cube, recover the unique underlying tree, and then recompute its cube to obtain the original graph.

Implications and Future Work (Section 7).
The result extends the class of graphs known to satisfy the Reconstruction Conjecture and demonstrates that the conjecture holds for a non‑trivial family of graphs that are not themselves trees but retain a tree‑like hierarchical structure. The authors suggest that similar techniques might be applicable to higher powers (fourth power, etc.) or to powers of graphs with bounded treewidth. They also discuss the computational aspects of the reconstruction algorithm, noting that the dominant step—identifying maximal cliques in each card—can be performed in (O(n^{3})) time, making the overall procedure feasible for moderate‑size graphs.

In summary, the paper provides a thorough and elegant proof that the Reconstruction Conjecture is true for graphs isomorphic to the cube of a tree, enriching our understanding of graph powers and offering new tools for tackling the conjecture in broader contexts.