Succinct indexes for circular patterns

Abstract

Circular patterns are those patterns whose circular permutations are also valid patterns. These patterns arise naturally in bioinformatics and computational geometry. In this paper, we consider succinct indexing schemes for a set of d circular patterns of total length n, with each character drawn from an alphabet of size σ. Our method is by defining the popular Burrows-Wheeler transform (BWT) on circular patterns, based on which we achieve succinct indexes with space n log σ(1 + o(1)) + O(n) + O(d log n) bits, while pattern matching or dictionary matching queries can be supported efficiently. © 2011 Springer-Verlag.