Assessing the stability of a phylogeny without bootstrap: an analytical approach

Assessing the stability of a phylogeny without bootstrap: an analytical approach Mahendra Mariadassou June 19, 2007 Contents 1 Introduction 2 2 Framework 3 2.1 Notations and definitions . . . . . . . . . . . . . . . . . . . . . . . . 3 2.2 Toy Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2.3 Likelihood and stability . . . . . . . . . . . . . . . . . . . . . . . . 5 3 Phylogenetic reconstruction for finite size samples 7 3.1 Connection between T and Q . . . . . . . . . . . . . . . . . . . . . 8 3.2 Distance between Q and Qn . . . . . . . . . . . . . . . . . . . . . . 8 3.3 Stability of the inferred tree . . . . . . . . . . . . . . . . . . . . . . 11 4 Comparison with widely used methods 13 4.1 Bootstrap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 4.2 Comparison on the toy example .

• aligned sequence


• aligned over


• species sites


• methods


• distance between


• change occurs only


• species called


• maximum likelihood


• sequence after


• species