On the application of evolutionary algorithms to the consensus tree problem

  • Authors:
  • Carlos Cotta

  • Affiliations:
  • Dept. Lenguajes y Ciencias de la Computación, University of Málaga, ETSI Informática, Málaga, Spain

  • Venue:
  • EvoCOP'05 Proceedings of the 5th European conference on Evolutionary Computation in Combinatorial Optimization
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

Computing consensus trees amounts to finding a single tree that summarizes a collection of trees. Three evolutionary algorithms are defined for this problem, featuring characteristics of genetic programming (GP), evolution strategies (ES) and evolutionary programming (EP) respectively. These algorithms are evaluated on a benchmark composed of phylogenetic trees computed from genomic data. The GP-like algorithm is shown to provide better results than the other evolutionary algorithms, and than two greedy heuristics defined ad hoc for this problem.