Elucidating the ecological, developmental, and environmental drivers of insect megadiversity

Project overview

Preventing the effects of human-caused habitat degradation and climate change requires detailed data on species biology in order to target conservation efforts.

Massive projects to date have focused on mapping where species live in order to predict how they will affected by environmental change. However, all organisms do not respond uniformly to changes in their environment.

To develop the best possible strategies to preserve biodiversity, we must know how species interact with their environment. For that, we need data on phenotype.

An organism's phenotype is how they interact with their environment and with other individuals, encompassing their anatomy, behaviour, and more. At present, however, collecting data on phenotype is largely manual and time-consuming, making it difficult to scale up to thousands of specimens.

Capturing phenotype is not just an issue of species numbers either; the incredible variation in the anatomy of organisms effectively prevent the use of existing approaches to compare very diverse groups. For example, 3D morphometrics has been applied to numerous groups of vertebrates, but rarely to the most diverse organisms on Earth: the insects.

In this project, newly developed mass scanning, deep-learning, computer vision, landmark-free morphometric methods, and multivariate evolutionary modelling will be applied to an unparalleled micro-CT dataset of insects to answer the fundamental but enigmatic question: why are there so many insects?

Project Specific Training

The student will receive training in cutting-edge approaches to imaging and image analysis, including use of the Diamond Light Source synchrotron, deep-learning and computer vision, morphometrics, and evolutionary modelling in R and Python.  Training will be primarily through one-to-one and group collaborative training in the supervisors’ labs, but also via external courses as needed.