Phylogenetic Comparative Methods and Tests of Macroevolutionary Questions

by

Emilia P. Martins, Jesualdo Arturo Fuentes-González

• LAST REVIEWED: 10 May 2023
• LAST MODIFIED: 26 April 2018
• DOI: 10.1093/obo/9780199941728-0099

Introduction

The comparative method has been used throughout biology to generate hypotheses and draw broad conclusions that apply to wide ranges of taxa. Typically, researchers will gather phenotypic data from several species (or other taxonomic groups) and then use statistical analyses of the species means to make historical inferences. They might ask, for example, whether two traits have been linked with each other over evolutionary time, whether the rate of phenotypic evolution differs across two clades or in two different types of traits, or whether one trait has been subject to greater phylogenetic constraint or inertia. Since the mid-1980s, it has been widely accepted that such analyses need to be conducted in a phylogenetic context, and a diversity of phylogenetic comparative methods (PCMs) have been proposed. From a statistical perspective, shared evolutionary history creates patterns (heteroscedasticity and interdependence) in a set of species means that violate the assumptions of parametric statistics and can lead to wrong answers. From an evolutionary perspective, only with a phylogeny can we truly understand the patterns and processes by which phenotypic evolution occurred. Using a phylogeny, researchers can ask important questions about macroevolutionary patterns and infer details of the evolutionary processes that led to phenotypic diversification. For example, by comparing differences in the distribution of phenotypes between clades or by comparing the inferred evolution of different types of traits, researchers can begin to infer the types and magnitudes of selection that have shaped phenotypic evolution. Often, researchers also explore relationships between phenotypes and environments as a way of testing for adaptation. As new data become easily accessible over the Internet, phylogenetic comparative methods have become increasingly important, helping to bring the modern evolutionary synthesis to fruition by offering ways to combine information from geology, geography, genomics, and organismal biology into a comprehensive reconstruction of trait evolution throughout the evolutionary past.

General Overview

There have been several reviews of phylogenetic comparative methods over the years, from a number of different perspectives. The field has been so active that reviews tend to be out of date before they are even published. Nevertheless, researchers can use the reviews to get a quick overview of the critical issues involved in analyzing their own data, and then rely on forward searches of the literature to ensure that they are applying the most recent advances. Felsenstein 2004 and O’Meara 2012 offer reviews in the broader context of phylogeny reconstruction, whereas Garamszegi 2014 brings a more practical approach to specific applications.

• Felsenstein, J. 2004. Inferring phylogenies. Sunderland, MA: Sinauer.

  A comprehensive review of phylogeny reconstruction and phylogenetic comparative methods. The book is thorough enough to serve as a reference, and it can also work well as a textbook.

• Garamszegi, L. Z., ed. 2014. Modern phylogenetic comparative methods and their application in evolutionary biology: Concepts and practice. Berlin: Springer-Verlag.

  An edited volume with chapters written by many of the key method developers. A practical introduction to how to implement phylogenetic comparative methods and the main issues to be considered along the way. Includes a website with tutorials and an active Facebook page for new developments.

• O’Meara, B. C. 2012. Evolutionary inferences from phylogenies: A review of methods. Annual Review of Ecology, Evolution, and Systematics 43.1: 267–285.

  DOI: 10.1146/annurev-ecolsys-110411-160331

  A review of phylogenetic methods that emphasizes underlying statistical properties, and that includes methods for phylogeny reconstruction as well as phylogenetic comparative methods. O’Meara divides phylogenetic methods into three types: continuous-time Markov chains with finite state space, multivariate normal distributions, and birth-death branching processes.