Abstract:
Tree-indexed Markov processes (and mixtures of such processes) are often used to model molecular sequence evolution. In this talk, we survey some recent results concerning the extent to which such models allow the underlying tree and associated parameters to be recovered from sequence data. In particular, we consider models that allow rate variation across sites, as well as a simple covarion model of site substitution. We also describe some new results concerning structural predictions on the shape of trees generated under simple speciation models. Such prior distribution trees are relevant to Bayesian approaches to molecular systematics.
Papers of interest:
Tuffley, C; Steel, M. 1998. Modeling the covarion hypothesis of nucleotide substitution. Mathematical Biosciences 147(1):63-91.
Steel, M; Szekely, L; Erdoes, PL; Waddell, P. 1993. A complete family of phylogenetic invariants for any number of taxa under Kimura's 3ST model. New Zealand Journal of Botany 31(3):289-296.