Invariants of phylogenies in a simple case with discrete states
- 195 Downloads
Under a simple model of transition between two states, we can work out the probabilities of different data outcomes in four species with any given phylogeny. For a given tree topology, if all characters are evolving under the same probabilistic model, there are two quadratic forms in the frequencies of outcomes that must be zero. It may be possible to test the null hypothesis that the tree is of a particular topology by testing whether these quadratic forms are zero. One of the tests is a test for independence in a simple 2×2 contingency table. If there are differences of evolutionary rate among characters, these quadratic forms will no longer necessarily be zero.
KeywordsPhylogenies Statistical tests
Unable to display preview. Download preview PDF.
- DAYHOFF, M.O., and ECK, R.V. (1968),Atlas of Protein Sequence and Structure 1967–1968, Silver Spring, Maryland: National Biomedical Research Foundation.Google Scholar
- FELLER, W. (1968),An Introduction to Probability Theory and Its Applications, Vol. I, 3rd ed., New York: Wiley.Google Scholar
- FELSENSTEIN, J. (1983a), “Inferring Evolutionary Trees from DNA Sequences,” inStatistical Analysis of DNA Sequence Data, Ed. B.S. Weir, New York: Marcel Dekker, 133–150.Google Scholar
- FELSENSTEIN, J. (1983b), “Statistical Inference of Phylogenies,”Journal of the Royal Statistical Society, A146, 246–272.Google Scholar
- FELSENSTEIN, J. (1985a), “Confidence Limits on Phylogenies with a Molecular Clock,”Systematic Zoology, 34, 152–161.Google Scholar
- FELSENSTEIN, J. (1985b), “confidence Limits on Phylogenies: An Approach Using the Bootstrap,”Evolution, 39, 783–791.Google Scholar
- NEYMAN, J. (1971), “Molecular Studies of Evolution: A Source of Novel Statistical Problems,” inStatistical Decision Theory and Related Topics, Eds. S.S. Gupta and J. Yackel, New York: Academic Press, 1–27.Google Scholar