The nucleus of a point determining graph

A graph is point determining if distinct points have distinct neighborhoods. In this paper we investigate the nucleus G⁰={υ?G:G–υ is point determining} of a point determining graph G. In particular, we characterize those graphs that are the nucleus of some connected point determining graph.     

On realizations of point determining graphs, and obstructions to full homomorphisms

A graph is point determining if distinct vertices have distinct neighbourhoods. A realization of a point determining graph H is a point determining graph G such that each vertex-removed subgraph G-x which is point determining, is isomorphic to H. We study the fine structure of point determining graphs, and conclude that every point determining graph has at most two realizations.A full homomorphism of a graph G to a graph H is a vertex mapping f such that for distinct vertices u and v of G, we have uv an edge of G if and only if f(u)f(v) is an edge of H. For a fixed graph H, a full H-colouring of G is a full homomorphism of G to H. A minimal H-obstruction is a graph G which does not admit a full H-colouring, such that each proper induced subgraph of G admits a full H-colouring. We analyse minimal H-obstructions using our results on point determining graphs. We connect the two problems by proving that if H has k vertices, then a graph with k+1 vertices is a minimal H-obstruction if and only if it is a realization of H. We conclude that every minimal H-obstruction has at most k+1 vertices, and there are at most two minimal H-obstructions with k+1 vertices.We also consider full homomorphisms to graphs H in which loops are allowed. If H has ? loops and k vertices without loops, then every minimal H-obstruction has at most (k+1)(?+1) vertices, and, when both k and ? are positive, there is at most one minimal H-obstruction with (k+1)(?+1) vertices.In particular, this yields a finite forbidden subgraph characterization of full H-colourability, for any graph H with loops allowed.     

Construction for bicritical graphs and k-extendable bipartite graphs

A graph G is said to be bicritical if G-u-v has a perfect matching for every choice of a pair of points u and v. Bicritical graphs play a central role in decomposition theory of elementary graphs with respect to perfect matchings. As Plummer pointed out many times, the structure of bicritical graphs is far from completely understood. This paper presents a concise structure characterization on bicritical graphs in terms of factor-critical graphs and transversals of hypergraphs. A connected graph G with at least 2k+2 points is said to be k-extendable if it contains a matching of k lines and every such matching is contained in a perfect matching. A structure characterization for k-extendable bipartite graphs is given in a recursive way. Furthermore, this paper presents an O(mn) algorithm for determining the extendability of a bipartite graph G, the maximum integer k such that G is k-extendable, where n is the number of points and m is the number of lines in G.     

Graphs with 1-Hamiltonian-connected cubes

The cube G³ of a connected graph G is that graph having the same vertex set as G and in which two distinct vertices are adjacent if and only if their distance in G is at most three. A Hamiltonian-connected graph has the property that every two distinct vertices are joined by a Hamiltonian path. A graph G is 1-Hamiltonian-connected if, for every vertex w of G, the graphs G and G?w are Hamiltonian-connected. A characterization of graphs whose cubes are 1-Hamiltonian-connected is presented.     

An upper bound on the size of a largest clique in a graph

A graph is point determining if distinct vertices have distinct neighborhoods. The nucleus of a point-determining graph is the set G^O of all vertices, v, such that G–v is point determining. In this paper we show that the size, ω(G), of a maximum clique in G satisfies ω(G) ? 2|π (G)^O|, where π(G) (the point determinant of G) is obtained from G by identifying vertices which have the same neighborhood.     

On a conjecture of Meyniel

A graph G is said to be very strongly perfect if for each induced subgraph H of G, each vertex of H belongs to a stable set that meets all maximal cliques of H. Meyniel proved that a graph is perfect if each of its odd cycles with at least five vertices contains at least two chords. Nowadays, such a graph is called a Meyniel graph. We prove that, as conjectured by Meyniel, a graph is very strongly perfect if and only if it is a Meyniel graph. We also design a polynomial-time algorithm which, given a Meyniel graph G and a vertex x of G, finds a stable set that contains x and meets all maximal cliques of G. We shall convert this algorithm into another polynomial-time algorithm which, given a Meyniel graph G, finds an optimal coloring of G, and a largest clique of G. Finally, we shall establish another property, related to perfection, of Meyniel graphs.     

Nuclei for totally point determining graphs

The nucleus (edge nucleus) of a point determining graph is defined by Geoffroy and Sumner to be the set of all points (edges) whose removal leaves the graph point determining. It is the purpose of this paper to develop the analogous concepts for totally point determining graphs, that is, graphs in which distinct points have distinct neighborhoods and closed neighborhoods.     

A well-quasi-order for tournaments

A digraph H is immersed in a digraph G if the vertices of H are mapped to (distinct) vertices of G, and the edges of H are mapped to directed paths joining the corresponding pairs of vertices of G, in such a way that the paths are pairwise edge-disjoint. For graphs the same relation (using paths instead of directed paths) is a well-quasi-order; that is, in every infinite set of graphs some one of them is immersed in some other. The same is not true for digraphs in general; but we show it is true for tournaments (a tournament is a directed complete graph).     

Path factors and parallel knock-out schemes of almost claw-free graphs

An H₁,{H₂}-factor of a graph G is a spanning subgraph of G with exactly one component isomorphic to the graph H₁ and all other components (if there are any) isomorphic to the graph H₂. We completely characterise the class of connected almost claw-free graphs that have a P₇,{P₂}-factor, where P₇ and P₂ denote the paths on seven and two vertices, respectively. We apply this result to parallel knock-out schemes for almost claw-free graphs. These schemes proceed in rounds in each of which each surviving vertex eliminates one of its surviving neighbours. A graph is reducible if such a scheme eliminates every vertex in the graph. Using our characterisation, we are able to classify all reducible almost claw-free graphs, and we can show that every reducible almost claw-free graph is reducible in at most two rounds. This leads to a quadratic time algorithm for determining if an almost claw-free graph is reducible (which is a generalisation and improvement upon the previous strongest result that showed that there was a O(n^5.376) time algorithm for claw-free graphs on n vertices).     

On clique-critical graphs

A graph G is clique-critical if G and G?x have different clique-graphs for all vertices x of G. For any graph H, there is at most a finite number of different clique-critical graphs G such that H is the clique-graph of G. Upper and lower bounds for the number of vertices of the cliques of the critical graphs are obtained.     

A common cover of graphs and 2-cell embeddings

Let G and H be finite graphs with equal uniform degree refinements. Their finite common covering graph G∘H is constructed. It is shown that G, H, and G∘H can be 2-cell embedded in orientable surfaces M, N and S^∼, respectively, in such a way that the graph covering projections G∘H → G and G∘H → H extend to branched coverings M ← S^∼ → N of the surfaces. Additional properties of G∘H are used to obtain some nontrivial consequences about coverings of some planar graphs.     

Constructing and classifying neighborhood anti-Sperner graphs

For a simple graph G let N_G(u) be the (open) neighborhood of vertex u∈V(G). Then G is neighborhood anti-Sperner (NAS) if for every u there is a v∈V(G)?{u} with N_G(u)⊆N_G(v). And a graph H is neighborhood distinct (ND) if every neighborhood is distinct, i.e., if N_H(u)≠N_H(v) when u≠v, for all u and v∈V(H).In Porter and Yucas [T.D. Porter, J.L. Yucas. Graphs whose vertex-neighborhoods are anti-sperner, Bulletin of the Institute of Combinatorics and its Applications 44 (2005) 69-77] a characterization of regular NAS graphs was given: ‘each regular NAS graph can be obtained from a host graph by replacing vertices by null graphs of appropriate sizes, and then joining these null graphs in a prescribed manner’. We extend this characterization to all NAS graphs, and give algorithms to construct all NAS graphs from host ND graphs. Then we find and classify all connected r-regular NAS graphs for r=0,1,…,6.     

Mutual exclusion scheduling with interval graphs or related classes. Part II

This paper is the second part of a study devoted to the mutual exclusion scheduling problem. Given a simple and undirected graph G and an integer k, the problem is to find a minimum coloring of G such that each color is used at most k times. The cardinality of such a coloring is denoted by χ(G,k). When restricted to interval graphs or related classes like circular-arc graphs and tolerance graphs, the problem has some applications in workforce planning. Unfortunately, the problem is shown to be NP-hard for interval graphs, even if k is a constant greater than or equal to four [H.L. Bodlaender, K. Jansen, Restrictions of graph partition problems. Part I. Theoret. Comput. Sci. 148 (1995) 93-109]. In this paper, the problem is approached from a different point of view by studying a non-trivial and practical sufficient condition for optimality. In particular, the following proposition is demonstrated: if an interval graph G admits a coloring such that each color appears at least k times, then χ(G,k)=⌈n/k⌉. This proposition is extended to several classes of graphs related to interval graphs. Moreover, all our proofs are constructive and provide efficient algorithms to solve the MES problem for these graphs, given a coloring satisfying the condition in input.     

Some perfect coloring properties of graphs

A Grundy n-coloring of a finite graph is a coloring of the points of the graph with the non-negative integers smaller than n such that each point is adjacent to some point of each smaller color but to none of the same color. The Grundy number of a graph is the maximum n for which it has a Grundy n-coloring. Characterizations are given of the families of finite graphs G such that for each induced subgraph H of G: (1) the Grundy number of H is equal to the chromatic number of H; (2) the Grundy number of H is equal to the maximum clique size of H; (3) the achromatic number of H is equal to the chromatic number of H; (4) the achromatic number of H is equal to the maximum clique size of H. The definitions are further extended to infinite graphs, and some of the above characterizations are shown to be true for denumerable graphs and locally finite graphs.     

Space graphs and sphericity

For a finite point set in Euclidean n-space, if we connect each pair of points by a line segment whenever the distance between them is less than a certain positive constant, we obtain a space graph in n-space. The sphericity of a graph G is defined to be the minimum number n such that G is isomorphic to a space graph in n-space. In this paper we study the sphericities of graphs and present upper bounds on the sphericity for several types of graphs.     

The edge nucleus of a point-determining graph

In this paper we investigate the edge nucleus E⁰(G) of a point-determining graph G. We observe several relationships between E⁰(G) and the nucleus G⁰ = {v ∈ V(G)∣ G ? v is point determining} and use these relationships to prove several properties of E⁰(G). In particular, we show that there are only a finite number of graphs with a given edge nucleus and we determine those graphs G for which |E⁰(G)| ≤ 2. We also show that an n-clique of a point-determining graph G contains at least n?2 edges of E⁰(G) and if G is totally point determining, then every odd cycle of G meets E⁰(G).     

On large light graphs in families of polyhedral graphs

A graph H is said to be light in a family H of graphs if each graph G∈H containing a subgraph isomorphic to H contains also an isomorphic copy of H such that each its vertex has the degree (in G) bounded above by a finite number φ(H,H) depending only on H and H. We prove that in the family of all 3-connected plane graphs of minimum degree 5 (or minimum face size 5, respectively), the paths with certain small graphs attached to one of its ends are light.     

Light graphs in families of polyhedral graphs with prescribed minimum degree, face size, edge and dual edge weight

A graph H is defined to be light in a family H of graphs if there exists a finite number φ(H,H) such that each G∈H which contains H as a subgraph, contains also a subgraph K≅H such that the Δ_G(K)≤φ(H,H). We study light graphs in families of polyhedral graphs with prescribed minimum vertex degree δ, minimum face degree ρ, minimum edge weight w and dual edge weight w^∗. For those families, we show that there exists a variety of small light cycles; on the other hand, we also present particular constructions showing that, for certain families, the spectrum of short cycles contains irregularly scattered cycles that are not light.     

On the intersection graph of a finite group

For a finite group G, the intersection graph of G which is denoted by Γ(G) is an undirected graph such that its vertices are all nontrivial proper subgroups of G and two distinct vertices H and K are adjacent when H ∩ K ≠ 1. In this paper we classify all finite groups whose intersection graphs are regular. Also, we find some results on the intersection graphs of simple groups and finally we study the structure of Aut(Γ(G)).