Research

To fully understand a biological system, we must take a deep look at how these systems function under normal conditions and when they are perturbed. Also important, is learning how the system can be modified to encode novel functions without detrimental effects – a fundamental process that leads to the amazing diversity of our natural world. 
 
Our research focuses on understanding animal behavior in the perspective of this evolvability: where and how are novel behavioral patterns generated at the genetic level and the neuronal level during evolution? What general principles may apply? Particularly, we are interested in the following questions:
 
– What kinds of genes and mutations are used to generate natural behavioral variations within and between species?
– How does the existing structure and function of neural circuits facilitate or constrain behavioral evolution?
– How converged are the genetic and neuronal solutions underlying repeated evolution of similar behavioral phenotypes?
 
Males belonging to Drosophila species perform courtship rituals that involve a suite of behaviors-  chasing the female, vibrating their wings to produce “song”, performing a dance, licking, and others. These behaviors evolve very rapidly and are essential for reproductive success and species recognition. Different species, even ones that are closely related, display rich variations in various aspects of courtship. Thanks to powerful genetic tools, the species D. melanogaster has been an excellent model system for genetics and neurobiology of various behaviors, including courtship. Applying these tools to related Drosophila species provides a unique opportunity to causally explain how behavior evolves at different levels of biological organization, from genes to neural circuits. Our lab uses a highly integrative approach spanning genetics, molecular biology and neurobiology, to address these questions. 
 

SongTorrent

(96-channel video and audio recording rig)

Homologous neurons in different species, projected to a standard template.