Investigating macroevolutionary diversification of feeding structures in sea urchins using deep learning

Abstract

Sea urchins (Echinoidea) have one of the most complete fossil records of any animal, presenting an ideal model system to answer a fundamental question in evolutionary biology: how does functional performance influence macroevolution across deep time? Crown group echinoids are characterized by a distinct structure – the perignathic girdle (PG) – which serves as the attachment site for the muscles that move their feeding apparatus. A wide diversity of PG morphologies exists, ranging from small, gracile pegs to robust wing-like ridges. While the PG has been recognised as a crucial feature underlying echinoid systematics, the functional roles of the different morphologies are unknown, making its role in echinoid diversification unclear. Combining deep learning approaches to microCT segmentation with 3D geometric morphometrics, we have quantified PG shape for 100 taxa, forming the largest growing databases of 3D echinoid morphology to date. Preliminary results reveal convergent evolution of PG shapes across evolutionarily distinct lineages, suggesting that there may be strong responses to selective pressures related to performance, with adaptive optima driving the evolution of different PG morphologies. This work provides insight into how changes in the functional performance of feeding morphology have influenced morphological and ecological diversification over the last 300 million years.

More information

Identifiers

ORCID for Alice Leavey: orcid.org/0000-0003-4062-8399

ORCID for Jeffrey R. Thompson: orcid.org/0000-0003-3485-172X

Catalogue record

Export record