The project deals with the development of a machine learning approach to characterise the molecular underpinning of social learning in honeybees during the waggle dance.
The waggle dance is a fascinating behaviour that honeybees use to communicate the location of a profitable food source to nestmates. This complex behaviour has been studied for long time and since the last century researchers have been successful in characterising all aspects of its regulation at the individual and group level. However, we don’t know much about its fine regulation at the molecular level.
In a previous project, we have started to investigate what genes in the brain of the honeybee might underpin the waggle dance communication. We have focused on two groups of bees: the bees that perform the dance (or ‘dancers’) to transfer the information they possess to nestmates, and the bees that attend the dance (or ‘dance followers’) interested in knowing the location of the new food site.
In this project, we use transcriptomic data obtained from dancers to implement a machine learning approach and test whether it is possible to predict the patterns of brain gene expression in followers that participated in the same dances. Machine learning approaches such as Support Vector Machine (SVM) have been used for different applications in biology, however, it has never been applied to understand how complex behaviour are regulated.
This is a unique opportunity to explore the molecular basis for social learning in a biological system that is an emerging model for the study of complex behaviours. Honeybees are key pollinators for wild flowers as well as for many agricultural crops, hence they provide key ecosystem services to both natural and human-dominated environments.
Understanding how the waggle dance communication system is regulated at the fine molecular scale will provide new tools to mediate the negative effects that environmental change is having on these important insects.