The availability of genome-wide within-species data leads to new opportunities in phylogenetics which is the study of evolutionary relationships of organisms.   DNA substitution models describe the evolution of DNA sequences as a series of substitution events which replace a nucleotide by another one.   Substitutions are modeled by a time-continuous, time-homogeneous Markov process.

 I present a maximum likelihood implementation of a reversible Polymorphism-Aware Phylogenetic Model (revPoMo) for species tree estimation which uses many within-species samples.   revPoMo expands the state space of DNA substitution models to include polymorphic states, thereby naturally accounting for incomplete lineage sorting.

I will explain how revPoMo mimics populations by separating substitutions into mutations with subsequent allele frequency shifts (genetic drift).   I will examine the stationary distribution of the underlying Markov process and compare it to the stationary solution of the Wright-Fisher diffusion equation.

 A simulation study and an application to great apes show that revPoMo is accurate in estimating species trees, divergence times and mutation rates.   The advantage of our approach is that an increase of sample size per species improves estimations but does not increase runtime.

Therefore, revPoMo is a valuable tool with several applications, from speciation dating to species tree reconstruction.