Phylogenetic Comparative Methods: A User’s Guide for Paleontologists

Abstract

Recent advances in statistical approaches called Phylogenetic Comparative Methods (PCMs) have provided paleontologists with a powerful set of analytical tools for investigating evolutionary tempo and mode in fossil lineages. However, attempts to integrate PCMs with fossil data often present workers with practical challenges or unfamiliar literature. In this paper, we present guides to the theory behind, and application of, PCMs with fossil taxa. Based on an empirical dataset of Paleozoic crinoids, we present example analyses to illustrate common applications of PCMs to fossil data, including investigating patterns of correlated trait evolution, and macroevolutionary models of morphological change. We emphasize the importance of accounting for sources of uncertainty, and discuss how to evaluate model fit and adequacy. Finally, we discuss several promising methods for modelling heterogenous evolutionary dynamics with fossil phylogenies. Integrating phylogeny-based approaches with the fossil record provides a rigorous, quantitative perspective to understanding key patterns in the history of life.

Microstructural and crystallographic evolution of palaeognath (Aves) eggshells

“…If some Late Cretaceous rhea-style ‘ratite-morphotype’ eggshells turn out to be true palaeognath eggshell, our interpretation (Figure 14) will be further supported with evidence. The ancestral state reconstruction of MD exemplifies the importance of fossils in ancestral reconstructions, especially when focusing on early nodes with a high discrepancy between extant and extinct species (e.g. Finarelli and Flynn, 2006; Li et al, 2008; Cascini et al, 2019; Soul and Wright, 2021). Maddison et al (1984) pointed out that, ideally, at least two outgroups are necessary to unambiguously polarize characters of ingroup taxa.…”

The diverse terminology of reptile eggshell microstructure and its effect on phylogenetic comparative analyses

“…ultrametric trees), in which no terminal taxon can have such a disproportionate influence. Including fossils in time‐calibrated trees for ASR requires precise evaluation of their phylogenetic position and fossil record sampling bias (Bapst, 2014), as well as their effect on the underlying evolutionary model of the trait of interest (Cascini et al, 2019; King & Lee, 2015; Litsios & Salamin, 2012; Wilson et al, in press) – both of which have been the subject of considerable debate in the paleontological community (Bapst, 2014; Soul & Wright, 2021). While the inclusion of both fossil and extant species generally improves ASR accuracy – especially when using discrete traits, which are less sensitive to sampling bias and model misspecification (Puttick, 2016; Soul & Wright, 2021), the effect of such a high discrepancy in distance between nodes and terminal taxa on ASR with time‐calibrated trees remains to be assessed.…”

“…In general, the field of ASR in paleontology is relatively new, and most biases associated with the inclusion of fossils for e.g. model fitting or sampling error have only been identified and characterized recently (Bapst, 2014; Hunt & Carrano, 2010; Soul & Wright, 2021). In this context, our results provide a clear example of inherent issues of uncertainty in tree topology and calibration, taxon sampling, and character coding (see next section), which are generally not considered, nor discussed, in recent studies on reptile eggshell evolution (e.g.…”

“…We therefore advocate for future studies performing ASR of eggshell type – or any other discrete or continuous egg trait – to prioritize, when possible, methods that can account for such biases most efficiently (e.g. hierarchical, split, or hidden rate Bayesian models; sensitivity analyses using joint estimation from a posterior distribution of trees – Joy et al, 2016; King & Lee, 2015; Soul & Wright, 2021; Wilson et al, in press; Wright et al, 2021), and explicitly discuss potential biases associated with these methods, sampled taxa, and data used for both traits and phylogeny tree used in such studies. These extra steps will help facilitate replication of previous results and improve future discussions on discrepancies between methods and datasets in this context.…”

“…scanning electron microscopy, EBSD) and chemical analysis (e.g. Raman spectroscopy, energy‐dispersive X‐ray spectroscopy, mass spectrometry) to describe eggshell microstructure and composition and measure quantitative traits; If analyzing a sample of eggs from different species, always correct for phylogenetic non‐independence (Paradis, 2012; Soul & Wright, 2021) and include eggs from every major clade in your group of interest, including fossils (e.g. a study on Reptilia should always sample avian eggs) – and, if possible, from a wide range of body sizes (e.g.…”