Evolutionary biology was defined by Charles Darwin, but he had his critics, such as Alfred Russel Wallace, who added the theory of evolution as we know it today.
Now, AI has contributed its own research to this debate thanks to a recent study from the University of Essex published in Communications Biology.
Darwin and Wallace debated what drives the remarkable diversity we observe in nature, particularly the striking differences between males and females of the same species.
Strong sexual dimorphism and dichromatism is often most prominent in insects and birds.
Darwin proposed that the often flamboyant appearance of male members from the same species resulted from sexual selection, with females choosing their mates based on aesthetic preferences.
Wallace, however, countered that natural selection, acting on both sexes, was the primary force shaping these differences.
Dr. Jennifer Hoyal Cuthill from the University of Essex and her team embarked on a mission to unravel the evolutionary secrets of stunning birdwing group butterflies
As Dr. Hoyal Cuthill explained in a blog post, “For the first time, we are able to measure the visible extents of evolution to test how much variation is present in different biological groups and among both males and females.”
How the study worked
The researchers used AI to analyze an impressive 16,000 photographs of birdwing butterflies, representing 35 species and 131 subspecies, from the Natural History Museum in London.
Here’s the process:
- Curation and digitization: The museum’s extensive birdwing butterfly collection was carefully curated and digitized, providing high-quality photographs of both male and female specimens.
- AI training: The researchers trained an AI algorithm called ButterflyNet to recognize and group images based on their visual similarities, enabling the AI to identify patterns and variations among the butterflies.
- Embedding and analysis: ButterflyNet generated a multidimensional “butterfly space” where similar images were clustered together, allowing the researchers to examine evolutionary relationships and differences between species and sexes.
- Genetic validation: To confirm the biological significance of the AI’s findings, the machine-learned embeddings were compared with genetic data, ensuring alignment with the evolutionary history of the birdwing butterflies.
- Quantifying sexual variation: The team developed a new measure called the “sexual disparity difference” to quantify the extent of variation between males and females, providing insights into the contributions of each sex to overall diversity.
The study’s findings supported both Darwin’s and Wallace’s theories.
Male birdwing butterflies generally exhibited more distinct shapes and patterns compared to females, aligning with Darwin’s idea of sexual selection.
However, the research also revealed substantial variation among females, particularly in the genus Troides.
Particularly, some female butterflies displayed greater diversity than their male counterparts, supporting Wallace’s hypothesis that natural selection can drive divergence in female phenotypes.
Dr. Hoyal Cuthill explained the findings: “Birdwings have been described as among the most beautiful butterflies in the world. This study gives us new insights into the evolution of their remarkable but endangered diversity.”
She adds, “High visible diversity among male butterflies supports the real-world importance of sexual selection from female mate choice on male variation, as originally suggested by Darwin. Cases where female butterflies are more visibly diverse than the males of their species, support an additional, important role for naturally selected female variation in inter-species diversity, as suggested by Wallace.”
This isn’t the first time AI has been applied to evolutionary biology. Not long ago, researchers from Colorado State University developed an AI system to explore how elephants communicate.
Researchers found that elephants call each other by name, similar to humans.
One of the cornerstones of Darwin’s theory of evolution was that animals, like humans, evolved skills like language for similar benefits to their survival.
The study’s authors described that elephants likely developed similar communication methods as humans due to their need to socialize and work together in groups.
As Dr. Hoyal Cuthill, from the School of Life Sciences, described: “This is an exciting time, when machine learning is enabling new, large-scale tests of longstanding questions in evolutionary science.”
