Inferring Evolutionary Timescales without Independent Timing Information: An Assessment of “Universal” Insect Rates to Calibrate a Collembola (Hexapoda) Molecular Clock

Abstract

Previous estimates of nucleotide substitution rates are routinely applied as secondary or “universal” molecular clock calibrations for estimating evolutionary timescales in groups that lack independent timing information. A major limitation of this approach is that rates can vary considerably among taxonomic groups, but the assumption of rate constancy is rarely evaluated prior to using secondary rate calibrations. Here I evaluate whether an insect mitochondrial DNA clock is appropriate for estimating timescales in Collembola—a group of insect-like arthropods characterized by high levels of cryptic diversity. Relative rates of substitution in cytochrome oxidase subunit 1 (COI) were inferred via Bayesian analysis across a topologically constrained Hexapod phylogeny using a relaxed molecular clock model. Rates for Collembola did not differ significantly from the average rate or from the rates estimated for most other groups (25 of 30), suggesting that (1) their apparent cryptic diversity cannot be explained by accelerated rates of molecular evolution and (2) clocks calibrated using “universal” insect rates may be appropriate for estimating evolutionary timescales in this group. However, of the 31 groups investigated, 10 had rates that deviated significantly from the average (6 higher, 4 lower), underscoring the need for caution and careful consideration when applying secondary insect rate calibrations. Lastly, this study exemplifies a relatively simple approach for evaluating rate constancy within a taxonomic group to determine whether the use of secondary rates are appropriate for molecular clock calibrations.