Efficient Estimation of the Accuracy of the Mazimum Likelihood Method for Reconstruction of Ancestral States
Ancestral state reconstruction is an important approach to investigating the evolution of molecular or morphological features of living organisms 1-5. Different reconstruction methods exist for ancestral state reconstruction from DNA or proteinsequences 6,7, multistate discrete data 8 10 and continuous data 11. These different methods have been studied analytically 12, 13 and empirically4,8,14-16. The reader is referred to the books of Felsenstein 17 and Liberles 18 for further information on ancestral state reconstruction. We are interested in reconstructing ancestral state from extant states, with the assumption that the true phylogeny that relates the extant taxa to the target ancestor is given. There are different methods for reconstructing ancestral states.Among them are the popular parsimony and marginal maximum likelihood (ML) methods. While the parsimony method 19 is efficient and works without a specific evolutionary model, the marginal (or local) maximum likelihood (ML) was proved to be the most accurate method once an evolutionary model is known （20, p.159 or 21, Theorem 4).
Bin Ma Louxin Zhang
David R.Cheriton School of Computer Science, University of Waterloo, Waterloo, Canada N2L 3G1 Department of Mathematics, National University of Singapore, Singapore 117543
国际会议
The 7th Asia-Pacific Bioinformatics Conference(第七届亚太生物信息学大会)
北京
英文
138-140
2009-01-01（万方平台首次上网日期，不代表论文的发表时间）