arXiv++ Combinatorics

Let $G$ be an $n$-vertex graph with adjacency matrix $A$, and $W=[e,Ae,\ldots,A^{n-1}e]$ be the walk matrix of $G$, where $e$ is the all-one vector. In Wang [J. Combin. Theory, Ser. B, 122 (2017): 438-451], the author showed that any graph $G$ is uniquely determined by its generalized spectrum (DGS) whenever $2^{-\lfloor n/2 \rfloor}\det W$ is odd and square-free. In this paper, we introduce a large family of graphs $\mathcal{F}_n=\{$ $n$-vertex graphs $G\colon\, 2^{-\lfloor n/2 \rfloor}\det W =p^2b$ and rank$W=n-1$ over $\mathbb{Z}/p\mathbb{Z}\},$ where $b$ is odd and square-free, $p$ is an odd prime and $p\nmid b$. We prove that any graph in $\mathcal{F}_n$ either is DGS or has exactly one generalized cospectral mate up to isomorphism. Moreover, we show that the problem of finding the generalized cospectral mate for a graph in $\mathcal{F}_n$ is equivalent to that of generating an appropriate rational orthogonal matrix from a given integral vector. This equivalence essentially depends on an amazing property of graphs in terms of generalized spectra, which states that any symmetric integral matrix generalized cospectral with the adjacency matrix of some graph must be an adjacency matrix. Based on this equivalence, we develop an efficient algorithm to decide whether a given graph in $\mathcal{F}_n$ is DGS and further to find the unique generalized cospectral mate when it is not. We give some experimental results on graphs with at most 20 vertices, which suggest that $\mathcal{F}_n$ may have a positive density (nearly $3\%$) and possibly almost all graphs in $\mathcal{F}_n$ are DGS as $n\rightarrow \infty$. This gives a supporting evidence for Haemers' conjecture that almost all graphs are determined by their spectra.

A pancyclic graph is a graph that contains cycles of all possible lengths from three up to the number of vertices in the graph. In this paper, we establish some new sufficient conditions for a graph to be pancyclic in terms of the edge number, the spectral radius and the signless Laplacian spectral radius of the graph.

Let $S_n(\mathbb{Z})$ and $O_n(\mathbb{Q})$ denote the set of all $n\times n$ symmetric matrices over the ring of integers $\mathbb{Z}$ and the set of all $n\times n$ orthogonal matrices over the field of rational numbers $\mathbb{Q}$, respectively. The paper is mainly concerned with the following problem: Given a matrix $A\in {S_n(\mathbb{Z})}$. How can one find all rational orthogonal matrices $Q\in{O_n(\mathbb{Q})}$ such that $Q^TAQ\in {S_n(\mathbb{Z})}$, and in particular, when does $Q^TAQ\in {S_n(\mathbb{Z})}$ with $Q\in{O_n(\mathbb{Q})}$ imply that $Q$ is \emph{a signed permutation matrix} (i.e., the matrix obtained from a permutation matrix $P$ by replacing each 1 in $P$ with 1 or $-1$)? A surprisingly simple answer was given in terms of whether the discriminant of the characteristic polynomial of $A$ is odd and square-free, which partially answers the above questions. More precisely, let $Δ_A=\pm \res(φ,φ')$ be \emph{the discriminant of matrix $A$}, where $\res(φ,φ')$ is \emph{the resultant} of the characteristic polynomial $φ$ of $A$ and its derivative $φ'$. We show that if $Δ_A$ is odd and square-free, then $Q^TAQ\in {S_n(\mathbb{Z})}$ with $Q\in{O_n(\mathbb{Q})}$ implies that $Q$ is a signed permutation matrix. As an application, we present a simple and efficient method for testing whether a graph is determined by the generalized spectrum, which significantly extends our previous work.