Phylogenetic Tip Rates: How Well Can We Estimate Diversification?

Post provided by Pascal Title and Dan Rabosky

Analyzing diversification rate heterogeneity across phylogenies allows us to explore all manner of questions, including why Australia has such an incredible diversity of lizards and snakes.

Analyzing diversification rate heterogeneity across phylogenies allows us to explore all manner of questions, including why Australia has such an incredible diversity of lizards and snakes.

Within the tree of life there are differences in speciation and extinction rates over time and across lineages. Biologists have long been interested in how speciation rates change as a function of ecological opportunity or whether key innovations lead to increases in the rate of speciation. Exploring this rate variation and examining how clades differ in terms of their diversification dynamics can help us to understand why species diversity varies so dramatically in time and space. Learning more about the relationship between traits and diversification rates is especially important because it has the potential to reveal the causes of pervasive variation in species richness among clades and across geographic regions.

Several different classes of methods are available for studying the effects of species traits on lineage diversification rates. These include state-dependent diversification models (e.g., BiSSE, QuaSSE, HiSSE) and several non-model-based approaches. In our article – ‘Tip rates, phylogenies and diversification: What are we estimating, and how good are the estimates?’ – we assessed the accuracy of a number of model-free metrics (the DR statistic, node density metric, inverse of terminal branch lengths) and model-based approaches (Bayesian Analysis of Macroevolutionary Mixtures, BAMM) to determine how they perform under a variety of different types of rate heterogeneity. The “tip rates” using these approaches have become widely used for a few reasons, including ease of computation and how easy it is to pair them with other types of data. Continue reading

Methods digest – update

A round up of recent methods-relevant research published recently: it is ages since we did this, largely because the journal has been so busy with papers coming in and being published. Do send through links to any new methods papers to me or to the journal, or post a comment below.

In Evolution, Werthelm & Sanderson look at how estimates of diversification rates are influenced by improved estimates of divergence times; Robert Lanfear introduces a new method for comparing rates of molecular evolution on trees.

In Systematic Biology Eric Stone has an extremely interesting article on why common comparative methods are robust to tree misspecification. Martin Linder et al. evaluate Bayesian models of substitution rate evoluton, whist Chung & Ané compare Bayesian methods for gene and species tree reconstructions. Simon Ho et al. have a short paper on Bayesian estimation of substitution rates from ancient DNA sequences.  Leaché & Rannala compare the accuracy of species tree estimation under different methods. Anne Kupczok explores the consequences of different null models for shape bias of supertree methods. John Huelsenbeck et al. compare phylogenetic models with the ‘No Common Mechanisms Model’.

In the Journal of Animal Ecology Andrew Jackson & co. have a paper on a new R package (SIBER) for comparing isotopic niche widths.

Sophie Smout et al. look at how heterogeneity of detection and mark loss affect estimates of survival in grey seals in Journal of Applied Ecology. Issue 1 of 2011 has a special profile introduced by Julia Jones on monitoring species abundance.

Eve McDonald-Madden et al. have a paper in Ecological Applications on how to allocate conservation resources when the persistence of a species in not certain. Mary Beth Rew and colleagues look at the problem of how many genetic markers should be used to tag an individual in the presence of close relatives.

A paper by Adam Algar et al. in Ecology looks at how it is possible to quantify the roles of trait-based filters in determining local and regional species composition. Florent Bled, Andy Royle & Emmanuelle Cam have a paper on testing hypotheses about nesting site dynamics by combining population and fitness data.

In Oikos, Sofia Berg et al. have a paper on the use of sensitivity analysis to identify keystones in foodwebs.

Finally for this update, in Ecography Simon Linke and co look at how multivariate analysis can produce conservation planning that addresses the needs of practitioners. Steinar Engen et al. describe a new approach to measuring the similarity of communities and Canrain Liu et al. have a paper on measuring the accuracy of species distribution models using presence absence data.

I’ll try to do another update in the next couple of weeks to cover some of the journals I have missed in this one.