Deep-learning-based pan-phenomic data reveals the explosive evolution of avian visual disparity

TL;DR

This study uses deep learning to analyze avian morphological evolution, revealing rapid diversification, phenotypic convergence, and the model's ability to learn holistic shape features beyond textures.

Contribution

It introduces a deep learning approach to quantify avian morphological disparity and demonstrates the model's capacity to encode biological taxonomy and shape-based features.

Findings

High-dimensional embedding encodes phenotypic convergence

Morphospace expansion driven mainly by species richness

Early burst of disparity after K-Pg extinction

Abstract

The evolution of biological morphology is critical for understanding the diversity of the natural world, yet traditional analyses often involve subjective biases in the selection and coding of morphological traits. This study employs deep learning techniques, utilising a ResNet34 model capable of recognising over 10,000 bird species, to explore avian morphological evolution. We extract weights from the model's final fully connected (fc) layer and investigate the semantic alignment between the high-dimensional embedding space learned by the model and biological phenotypes. The results demonstrate that the high-dimensional embedding space encodes phenotypic convergence. Subsequently, we assess the morphological disparity among various taxa and evaluate the association between morphological disparity and species richness, demonstrating that species richness is the primary driver of morphospace expansion. Moreover, the disparity-through-time analysis reveals a visual "early burst" after the K-Pg extinction. While mainly aimed at evolutionary analysis, this study also provides insights into the interpretability of Deep Neural Networks. We demonstrate that hierarchical semantic structures (biological taxonomy) emerged in the high-dimensional embedding space despite being trained on flat labels. Furthermore, through adversarial examples, we provide evidence that our model in this task can overcome texture bias and learn holistic shape representations (body plans), challenging the prevailing view that CNNs rely primarily on local textures.