We are reverse-engineering the fly, Drosophila melanogaster, to understand how animals generate flexible motor behaviors, leverage social information, and learn about the world. We believe that our efforts will uncover general insights into biological intelligence and can inform the design of better artificial systems and robots.

Flies are ideal for this goal: they generate complex behaviors yet have a small nervous system and are genetically malleable. For our research, we develop and use a variety of approaches including microscopy, machine learning, genetics, and computational modeling. We are part of the Brain Mind Institute and Institute of Bioengineering in the School of Life Sciences at EPFL, Switzerland 🇨🇭.

• Uncovering information flow between the brain and motor system (genetics, 2-photon imaging, machine learning-based image analysis) Relevant publications: Braun, Hurtak et al., Nature 2024, Chen et al., Nature Neuroscience, 2023, Aymanns et al., Elife, 2022;

• Building a neuromechanical model of Drosophila in a physics environment (physics simulations, biomechanics, machine learning) Relevant publications: Lobato et al., Nature Methods, 2022, Wang-Chen et al., bioRxiv, 2023

• Recording neural activity in the motor system during behavior (2-photon imaging, microengineering) Relevant publications: Hermans, Kaynak et al., Nature Communications 2022, Chen, Hermans et al., Nature Communications 2018

• Using deep networks to efficiently and precisely quantify behavior (machine learning, computer vision) Relevant publications: Gosztolai, Günel et al. Nature Methods 2021, Günel et al. Elife 2019, Günel et al. IJCV 2023

• Robotic experimental automation for high-throughput behavioral experiments (robotics, computer vision) Relevant publications: Ramdya et al., Nature 2015, Maesani, Ramdya et al., PLoS Computational Biology 2015

For further information, see our Publications page.