New paper about using morphometrics for phylogenetic inference (in press in Systematic Biology)

Out as “advance article” in Systematic Biology a new open access paper on using morphometric data in phylogenetic inference.

The work in the paper has been carried out for the most part in the lab of Matt Phillips at Queensland University of Technology (Australia). The work started during the time Carmelo was a postdoc in the Phillips lab at QUT, as part of the PhD work of Melina Celik (lead author in the paper) and Carmelo – who is a coauthor in the study – has contributed mainly to the morphometric part both at that time and more recently when the manuscript was finalised.

The paper deals with how to try improving the “objectivity” of morphological phylogenetics (which is quite important with fossil data and total evidence phylogenetic analysis) using geometric morphometric data. The core idea of the paper is that, rather than using geometric morphometric data for an anatomical structure as continuous data in phylogenetic inference or using traditional multi-state morphological characters which are not always scored in an objective manner, one can first subdivide an anatomical structure in functional portions and then discretise the variation in the morphometric data corresponding to each portion to obtain more objective multi-state characters. This core idea can be, obviously, applied in many ways and so the paper tests several approaches, for example for how to discretise morphometric variation. The results show that this approach performs well compared to competing approaches (such as some that analyse continuous morphometric variation directly) and produce reasonable phylogenies compared to the molecular phylogenetic benchmark (as most of the problems occur at poorly resolved parts of the phylogeny used as reference).