A molecular timescale for galliform birds accounting for uncertainty in time estimates and heterogeneity of rates of DNA substitutions across lineages and sites

Abstract

A recent molecular timescale for major lineages of the Galliformes indicated that Megapodiidae and possibly Cracidae, originated in the Cretaceous, while the remaining families originated in the Tertiary. This timescale was based on clock-like evolution in genetic and taxonomic partitions of mitochondrial ND2 and cyt b DNA sequences, and assumed that ordinal diversification of Galloanserae around 90 million years ago and imposed, whenever appropriate, minimum age constraints based on the fossil record. This approach is not ideal, as it did not account for uncertainty in estimating branch lengths and time, including the calibration time, and heterogeneity in the rate of DNA substitution among sites and in different lineages. Furthermore all the information available in the DNA sequences was not included, and may have been affected by stochastic error in individual gene partitions. Here, we present a follow-up analysis by estimating divergence times using a Bayesian framework that accounts for these possible sources of uncertainty. Our results based on combined and separate analyses of mitochondrial DNA sequences comprised of 1756 sites of 12S rDNA, ND2 and cyt b indicated that (1) Megapodiidae and Cracidae, and likely Odontophoridae, originated in the Cretaceous; (2) estimates based on concatenated genes are less affected by stochastic error among sites and less influenced by the phylogenetic signals of individual gene partitions, which are unequally distributed along the phylogenetic tree; and (3) the use of only an external molecular calibration results in lower estimation of most ingroup node ages. We also point out that galliform fossils may not be as useful for point calibrations as was previously suggested, but instead may be better employed as priors for the estimation of node ages under a Bayesian approach.