The merging of succinct data structures is a well established technique for the space efficient construction of large succinct indexes. In the first part of the paper we propose a new algorithm for merging succinct representations of de Bruijn graphs. Our algorithm has the same asymptotic cost of the state of the art algorithm for the same problem but it uses less than half of its working space. A novel important feature of our algorithm, not found in any of the existing tools, is that it can compute the Variable Order succinct representation of the union graph within the same asymptotic time/space bounds. In the second part of the paper we consider the more general problem of merging succinct representations of Wheeler graphs, a recently introduced graph family which includes as special cases de Bruijn graphs and many other known succinct indexes based on the BWT or one of its variants. We show that Wheeler graphs merging is in general a much more difficult problem, and we provide a space efficient algorithm for the slightly simplified problem of determining whether the union graph has an ordering that satisfies the Wheeler conditions.
翻译:简洁数据结构的合并是空间高效构建大型简洁索引的公认技术。在论文第一部分,我们建议采用新的算法,将德布鲁日图的简洁表述合并。我们的算法对同一问题有着同样的算法状态的微量成本,但它使用不到其工作空间的一半。我们算法中的一个重要新颖特征是,它可以在相同的时间/空间界限内对组合图的可变顺序简洁表述进行计算。在论文第二部分,我们考虑了将惠勒图的简洁表述合并这一更为普遍的问题。最近推出的图表系列包括德布鲁日图表的特殊案例和许多其他基于BWT或其变式之一的已知简洁索引。我们表明,惠勒图的合并一般是一个困难得多的问题,我们为确定工会图表是否具有符合惠勒号条件的简单问题提供了一种空间高效的算法。