Solving 3-superstring in 3n/3 time

5Citations
Citations of this article
2Readers
Mendeley users who have this article in their library.
Get full text

Abstract

In the shortest common superstring problem (SCS) one is given a set s 1,..., sn of n strings and the goal is to find a shortest string containing each si as a substring. While many approximation algorithms for this problem have been developed, it is still not known whether it can be solved exactly in fewer than 2n steps. In this paper we present an algorithm that solves the special case when all of the input strings have length 3 in time 3n/3 and polynomial space. The algorithm generates a combination of a de Bruijn graph and an overlap graph, such that a SCS is then a shortest directed rural postman path (DRPP) on this graph. We show that there exists at least one optimal DRPP satisfying some natural properties. The algorithm works basically by exhaustive search, but on the reduced search space of such paths of size 3n/3. © 2013 Springer-Verlag.

Cite

CITATION STYLE

APA

Golovnev, A., Kulikov, A. S., & Mihajlin, I. (2013). Solving 3-superstring in 3n/3 time. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8087 LNCS, pp. 480–491). https://doi.org/10.1007/978-3-642-40313-2_43

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free