Extension and use of KMRCRelat algorithm for biological problems
Issue title: Selected papers from the International Conference on Computer Science, Software Engineering, Information Technology, e-Business, and Applications, 2004
Affiliations: [a] LIPN, UMR-CNRS 7030, Université Paris-Nord, 93430 Villetaneuse, France | [b] L.E.R.I.A., 14/16 Rue Voltaire, 94270 Kremlin Bicetre, France. E-mail: el-kad_n@epitech.net | [c] INSERM Unit 558 faculty of Medecine 37, allées J. Guesde F-31073, Toulouse Cedex, France. E-mail: gourraud@cict.fr
Correspondence:
[*]
Corresponding author: Nahla EL Zant EL Kadhi, LIPN Institut Gallile. Tel.: +33614418729; Fax: +33144080120; E-mail: znahla@club-internet.fr.
Abstract: The main goal of this paper is to present a new extension of KMRCRelat algorithm allowing Word Train searches. First, we recall the fundamental lemma of our KMRCRelat algorithm and we focus on the concept of flexible relational repeated words. Then, we introduce the concept of Word Trains and we show, in deep, the needed modification and extension to include such features in this algorithm. To illustrate Word Train searches, we present all details about KMRCRelat extension computation steps. We also show how this new extension is applicable to the characterization of short tandem repeat in DNA sequence providing an alternative way to solve originally this known problem in sequence analysis. Before concluding by introducing other possible KMRCRelat applications, the paper expands a set of experimental results obtained when applied to locus identification problem.
Keywords: Pattern recognition, biological sequences, relational patterns, microsatellite identification, word train, words, patterns, sequences, repetitions, locus identification
