Constructions of (q,k,1) difference families with q a prime power and k=4,5
Selected papers of the 14th British conference on Combinatorial conference
On the number of blocks in a generalized Steiner system
Journal of Combinatorial Theory Series A
Pairwise balanced designs from finite fields
Discrete Mathematics
Two New Quorum Based Algorithms for Distributed Mutual Exclusion
ICDCS '97 Proceedings of the 17th International Conference on Distributed Computing Systems (ICDCS '97)
On spaced seeds for similarity search
Discrete Applied Mathematics
IEEE/ACM Transactions on Computational Biology and Bioinformatics (TCBB)
Linear work suffix array construction
Journal of the ACM (JACM)
On the complexity of the spaced seeds
Journal of Computer and System Sciences
Optimal spaced seeds for faster approximate string matching
Journal of Computer and System Sciences
Hardness of optimal spaced seed design
Journal of Computer and System Sciences
Constructions for strictly cyclic 3-designs and applications to optimal OOCs with λ=2
Journal of Combinatorial Theory Series A
ZOOM! Zillions of oligos mapped
Bioinformatics
Bioinformatics
Bioinformatics
Bioinformatics
Better Filtering with Gapped q-Grams
Fundamenta Informaticae - Computing Patterns in Strings
| Hi-index | 5.23 |
A wide class of approximate pattern matching algorithms are based on a filtration phase in which spaced seeds are used to discard regions where a match is not likely to occur. The problem of determining the ''optimal'' shape of a spaced seed in a particular setting is known to be a hard one: in practice spaced seeds are chosen using heuristics or considering a restricted family of seeds with ''reasonably good'' performances. In this paper we consider the family of spaced seeds with a periodic structure. Such seeds have been already proven valuable both as a theoretical tool and in bioinformatics applications. We show that known combinatorial objects, namely Difference Sets and Families and Steiner Systems, naturally lead to the design of lossless periodic seeds for approximate pattern matching with k=2 and k=3 mismatches. We analyze in depth the properties of the resulting seeds obtaining insights also on seeds without a periodic structure. The results of the analysis are then used to guide an experimental evaluation of the effectiveness of periodic seeds for pattern lengths of practical interest. Our results give a complete picture of strengths and limitations of periodic seeds, and can be used by practitioners for the design of effective approximate pattern matching algorithms.