Dive Deep into Probabilistic Phylogenetic Comparative Methods

This course offers an advanced understanding of probabilistic inference and its application for Phylogenetic Comparative Methods (PCM). Participants will gain a deeper knowledge of the stochastic processes, their inference and computation behind PCMs as well as their biological interpretations.

We will dive into probabilistic inference, first using Maximum Likelihood and, secondly, within a Bayesian framework, reviewing basic probability concepts and their application to posterior parameter estimation. Participants will learn how to perform inference from scratch (design the likelihood function, find Maximum Likelihood Estimates, implement and run an MCMC chain). Finally, most of the course will then delve into the main three PCM: trait and biogeographic evolution, and a deeper emphasis on diversification models. Topics covered include basic foundations (i.e., diffusion processes such as Brownian motion, time-continuous Discrete Markov models, birth-death models) to then build-up to the more advanced models that allow for interdependence between processes (i.e., environmental and geographic diversification, inference of biotic interactions). The course will combine introductory lectures and hands-on exercises.

We will start with an introduction to the Julia language (a powerful language for numerical computing that combines high performance with accessible high-level language), and use both R and Julia simultaneously when explaining probabilistic inference, assuring a smooth transition to the use of PCMs in Julia in the second part of the course. Previous experience in Julia is not needed.

Places are limited to 18 participants.