Computing bond orders in molecule graphs

In this paper, we deal with restoring missing information in molecule databases: Many data formats only store the atoms’ configuration but omit bond multiplicities. As this information is essential for various applications in chemistry, we consider the problem of recovering bond type information using a scoring function for the possible valences of each atom—the Bond Order Assignment problem. We show that the Bond Order Assignment is NP-hard, and its maximization version is MAX SNP-hard. We then give two exact fixed-parameter algorithms for the problem, where bond orders are computed via dynamic programming on a tree decomposition of the molecule graph. We evaluate our algorithm on a set of real molecule graphs and find that it works fast in practice.