The Community for Technology Leaders
RSS Icon
Subscribe
Issue No.01 - January-March (2009 vol.6)
pp: 126-133
Elchanan Mossel , UC Berkeley, Berkeley
Sebastien Roch , Microsoft Research, Redmond
Mike Steel , University of Canterbury, Christchurch
ABSTRACT
Ancestral maximum likelihood (AML) is a method that simultaneously reconstructs a phylogenetic tree and ancestral sequences from extant data (sequences at the leaves). The tree and ancestral sequences maximize the probability of observing the given data under a Markov model of sequence evolution, in which branch lengths are also optimized but constrained to take the same value on any edge across all sequence sites. AML differs from the more usual form of maximum likelihood (ML) in phylogenetics because ML averages over all possible ancestral sequences. ML has long been know to be statistically consistent - that is, it converges on the correct tree with probability approaching 1 as the sequence length grows. However, the statistical consistency of AML has not been formally determined, despite informal remarks in a literature that dates back 20 years. In this short note we prove a general result that implies that AML is statistically inconsistent. In particular we show that AML can 'shrink' short edges in a tree, resulting in a tree that has no internal resolution as the sequence length grows. Our results apply to any number of taxa.
INDEX TERMS
Markov processes, Biology and genetics
CITATION
Elchanan Mossel, Sebastien Roch, Mike Steel, "Shrinkage Effect in Ancestral Maximum Likelihood", IEEE/ACM Transactions on Computational Biology and Bioinformatics, vol.6, no. 1, pp. 126-133, January-March 2009, doi:10.1109/TCBB.2008.107
REFERENCES
[1] L. Addario-Berry, B. Chor, M. Hallett, J. Lagergren, A. Panconesi, and T. Wareham, “Ancestral Maximum Likelihood of Evolutionary Trees Is Hard,” J. Bioinformatics and Computational Biology, vol. 2, no. 2, pp. 257-271, 2004.
[2] N. Alon, B. Chor, F. Pardi, and A. Rapoport, “Approximate Maximum Parsimony and Ancestral Maximum Likelihood,” IEEE/ACM Trans. Computational Biology and Bioinformatics, to appear.
[3] D. Barry and J. Hartigan, “Statistical Analysis of Hominoid Molecular Evolution,” Statistical Science, vol. 2, pp. 191-207, 1987.
[4] D. Barry and J. Hartigan, “Rejoinder [On Statistical Analysis of Hominoid Molecular Evolution],” Statistical Science, vol. 2, pp.209-210, 1987.
[5] J.A. Cavender, “Taxonomy with Confidence,” Math. Biosciences, vol. 40, nos. 3/4, 1978.
[6] B. Chor and T. Tuller, “Finding the Maximum Likelihood Tree Is Hard,” Proc. Ninth Ann. Int'l Symp. Research in Computational Biology (RECOMB), 2005.
[7] P.L. Erdös, M.A. Steel, L.A. Székely, and T. Warnow, “A Few Logs Suffice to Build (Almost) All Trees (Part 1),” Random Structures and Algorithms, vol. 14, no. 2, pp. 153-184, 1999.
[8] J.S. Farris, “A Probability Model for Inferring Evolutionary Trees,” Systematic Zoology, vol. 22, no. 4, pp. 250-256, 1973.
[9] J. Felsenstein, “Cases in Which Parsimony or Compatibility Methods Will Be Positively Misleading,” Systematic Biology, vol. 27, pp. 401-410, 1978.
[10] J. Felsenstein, Inferring Phylogenies. Sinauer, 2004.
[11] N. Goldman, “Maximum Likelihood Inference of Phylogenetic Trees, with Special Reference to a Poisson Process of DNA Substitution and to Parsimony Analysis,” Systematic Zoology, vol. 39, pp. 345-361, 1990.
[12] P.A. Goloboff, “Parsimony, Likelihood, and Simplicity,” Cladistics, vol. 19, pp. 91-103, 2003.
[13] V. Moulton and M. Steel, “Peeling Phylogenetic ‘Oranges’,” Advances in Applied Math., vol. 33, no. 4, pp. 710-727, 2004.
[14] J. Neyman, “Molecular Studies of Evolution: A Source of Novel Statistical Problems,” Statistical Decision Theory and Related Topics, S.S. Gupta and J. Yackel, eds., pp. 1-27, Academic Press, 1971.
[15] T. Pupko, I. Pe'er, R. Shamir, and D. Graur, “A Fast Algorithm for Joint Reconstruction of Ancestral Amino Acid Sequences,” Molecular Biology and Evolution, vol. 17, no. 6, pp. 890-896, 2000.
[16] P. Ralph, in preparation, 2008.
[17] S. Roch, “A Short Proof That Phylogenetic Tree Reconstruction by Maximum Likelihood Is Hard,” IEEE/ACM Trans. Computational Biology and Bioinformatics, vol. 3, no. 1, pp. 92-94, Jan.-Mar. 2006.
22 ms
(Ver 2.0)

Marketing Automation Platform Marketing Automation Tool