Welcome to the InfoSci Platform
Reference Hub6
De novo Motif Prediction using the Fireworks Algorithm

De novo Motif Prediction using the Fireworks Algorithm

Andrei Lihu, Ștefan Holban
Copyright: © 2015 |Volume: 6 |Issue: 3 |Pages: 17
ISSN: 1947-9263|EISSN: 1947-9271|EISBN13: 9781466678293|DOI: 10.4018/IJSIR.2015070102
Cite Article Cite Article

MLA

Lihu, Andrei, and Ștefan Holban. "De novo Motif Prediction using the Fireworks Algorithm." IJSIR vol.6, no.3 2015: pp.24-40. http://doi.org/10.4018/IJSIR.2015070102

APA

Lihu, A. & Holban, Ș. (2015). De novo Motif Prediction using the Fireworks Algorithm. International Journal of Swarm Intelligence Research (IJSIR), 6(3), 24-40. http://doi.org/10.4018/IJSIR.2015070102

Chicago

Lihu, Andrei, and Ștefan Holban. "De novo Motif Prediction using the Fireworks Algorithm," International Journal of Swarm Intelligence Research (IJSIR) 6, no.3: 24-40. http://doi.org/10.4018/IJSIR.2015070102

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

De novo motif discovery is essential in understanding the cis-regulatory processes that play a role in gene expression. Finding unknown patterns of unknown lengths in massive amounts of data has long been a major challenge in computational biology. Because algorithms for motif prediction have always suffered of low performance issues, there is a constant effort to find better techniques. Evolutionary methods, including swarm intelligence algorithms, have been applied with limited success for motif prediction. However, recently developed methods, such as the Fireworks Algorithm (FWA) which simulates the explosion process of fireworks, may show better prospects. This paper describes a motif finding algorithm based on FWA that maximizes the Kullback-Leibler divergence between candidate solutions and the background noise. Following the terminology of FWA's framework, the candidate motifs are fireworks that generate additional sparks (i.e. derived motifs) in their neighborhood. During the iterations, better sparks can replace the fireworks, as the Fireworks Motif Finder (FW-MF) assumes a one occurrence per sequence mode. The results obtained on a standard benchmark for promoter analysis show that our proof of concept is promising.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.