Molecular Clocks without Rocks: New Solutions for Old Problems. Issue 11 (November 2020)

• Record Type: Journal Article
• Authors: Tiley, George P.; Poelstra, Jelmer W.; dos Reis, Mario; Yang, Ziheng; Yoder, Anne D.
• Is Part Of: Trends in genetics. Volume 36:Issue 11(2020)
• Page Start: 845
• Page End: 856
• Publication Date: 2020-11
• Subjects: divergence time estimation -- multispecies coalescent -- substitution rate -- mutation rate -- effective population size -- gene tree discordance; Genetics -- Periodicals; 576.5
• DOI: 10.1016/j.tig.2020.06.002
• Languages: English
• ISSNs: 0168-9525

Abstract

Abstract : Molecular data have been used to date species divergences ever since they were described as documents of evolutionary history in the 1960s. Yet, an inadequate fossil record and discordance between gene trees and species trees are persistently problematic. We examine how, by accommodating gene tree discordance and by scaling branch lengths to absolute time using mutation rate and generation time, multispecies coalescent (MSC) methods can potentially overcome these challenges. We find that time estimates can differ – in some cases, substantially – depending on whether MSC methods or traditional phylogenetic methods that apply concatenation are used, and whether the tree is calibrated with pedigree-based mutation rates or with fossils. We discuss the advantages and shortcomings of both approaches and provide practical guidance for data analysis when using these methods. Highlights: Molecular clock models using fossil calibrations have allowed investigators to estimate the age of speciation events. Theoretical and computational developments have relaxed the assumption of a molecular clock, thus improving the accuracy of divergence time estimation. Despite these advances, estimates can be biased when there is widespread incomplete lineage sorting (ILS). Increased understanding of gene tree heterogeneity has driven multispecies coalescent (MSC) methods to prominence, though the potential power of the MSC for divergence time estimation remains largely unexplored. Absolute times can be obtained by using mutation rates estimated from pedigrees, providing (some) freedom from the incomplete fossil record. Mutation-rate calibrated MSC methods and traditional phylogenetic clock-dating methods with fossil calibrations can yield strikingly different divergence times.