Class DistanceTreeEvaluator


  • public class DistanceTreeEvaluator
    extends Object
    Check the accuracy of a Distance Tree by least squares error (LSE) of the Tree branch lengths and the original Distance Matrix.
    Author:
    Scooter Willis, Aleix Lafita
    • Method Detail

      • evaluate

        public static double evaluate​(org.forester.phylogeny.Phylogeny tree,
                                      org.forester.evoinference.matrix.distance.DistanceMatrix matrix)
        Evaluate the goodness of fit of a given tree to the original distance matrix. The returned value is the coefficient of variation, i.e. the square root of the LS error normalized by the mean.

        This measure can also give an estimate of the quality of the distance matrix, because a bad fit may mean that the distance is non-additive.

        Parameters:
        tree - Phylogenetic Distance Tree to evaluate
        matrix - Distance Matrix with the original distances
        Returns:
        the square root of the average tree LS error normalized by the average tree distance (coefficient of variation, CV).