PhD: Integrating morphological and molecular methods of telling evolutionary time

Traditionally, the fossil record has provided the timescale for evolutionary history. However, some evolutionary lineages are completely unrepresented in the fossil record and the rock record provides an inconsistent sampling of environments through Earth History. Given the frailties of the fossil record, molecular clocks have become the tool of choice in attempts to tell evolutionary time. However, there has been a sea-change in perspective among molecular biologists, who are once again looking to morphological data and the fossil record to better inform the models and assumptions on which their molecular analyses are based, with the ultimate aim of developing a reliable means of obtaining an evolutionary timescale. Unfortunately, the single existing model of morphological trait data are naïve and under-developed.

The focus of this studentship project will be on the development and implementation of new models of morphological evolution within phylogenetic and divergence dating analyses. In particular, existing models make no account of the contingent relations and correlations that are known to exist among characters. In an approach akin to the development of empirical amino acid substitution models, the student will first analyse divers
e empirical datasets and both the theoretical and practical relationships among characters. They will then develop the existing likelihood model to accommodate the general qualities of empirical data and evaluate performance using a simulation approach. The model will be refined and used in the analysis of empirical datasets chosen to address high profile controversies in evolutionary biology.

Ultimately, model development is crucial not only to the establishment of timescales for studying evolutionary history. Models of morphological trait evolution, such as models of drift in landmark coordinate in bones, are mathematically the same as those used to deal with trait evolution in viruses, such as drift models of geographical coordinates of virus spread, or flux in the patient’s CD4 count, where sampling dates act as calibrations. Thus, the models developed and training gained through this research project have much broader potential.
This is a unique opportunity for a talented palaeobiologist, comparative anatomist, computational or molecular biologist, to work within an established collaborative team that includes expertise in the development of Bayesian molecular clock methods, in the implementation of morphological data into molecular clock analyses, as well as in the nature of palaeontological, morphological, molecular and phylogenetic data.