R Hackathon 1/Trait Evolution SG

From Phyloinformatics
Revision as of 16:31, 24 January 2008 by Lukeh@uidaho.edu (talk)
Jump to: navigation, search
  • Participants: Harmon, Hipp, Hunt


  1. Compare various implementations of the same methods (ape, geiger, OUCH, Mesquite)
  2. Improve functionality of character fitting in r
  3. Investigate statistical power of discriminating among various models
  4. Identify gaps in current implementation


  1. Evaluated the results of continuous character analyses in different packages
    • Packages are mostly consistent
    • Discrepancies come from two sources:
      • Different approaches (e.g. marginal versus joint likelihood)
      • Difficulties in finding the ML solution
    • For continuous characters:
      • geiger and OUCH tend to return the same parameter estimates
      • But they return different likelihoods
    • For discrete characters
      • geiger and mesquite are consistent, returning the same parameter estimates and likelihoods
      • geiger and ape are different
      • ape is reporting the joint likelihoods for ancestral states. This represents the single set of ancestral

states that together result in the highest likelihood on the whole tree. mesquite and geiger use marginal likelihoods for ancestral states. This represents the likelihood averaged over all possible ancestral character state values.

  1. Improve functionality or, at least, interpretability of output
    • GEIGER was modified to give more reliable results by a more thorough search of the likelihood surface
    • Some ape functions seem (to us) unreliable for large trees


    • Clarified why one sometimes gets different results from different programs
    • Modified GEIGER package so that results for fitting models of character evolution are more robust
    • Modified functions to rescale trees and variance-covariance matrices for non-ultrametric trees