Genetic algorithms and simulated annealing for gene mapping

Ordering genes on a chromosome is similar to the traveling salesman problem (TSP), but with some extra information. This extra information prompted us to investigate the genetic algorithms (GA) and simulated annealing (SA) methods to solve the gene ordering problem, even though they are not the best methods available to solve the TSP. Comparing the two heuristics on the gene mapping problem, we find that the GA method always converges to a good solution more quickly than the SA method. The best solution produced by the GA method is always superior to the best solution produced by the SA method. The GA method is able to do well on this problem since it is able to take advantage of the extra information mentioned above to construct good local maps that can then be used to construct good global maps.<<ETX>>