Karen Y. Cheng

The paper develops a deep-learning framework to learn interpretable representations of whole-brain neural dynamics in Drosophila without neuronal identification, revealing factorized organization along metabolic state, sensory modality, and stimulus valence axes.

The paper examines brain activity during walking in Drosophila, finding global but heterogeneous patterns of neuronal activity associated with both spontaneous and forced walking.

This paper demonstrates that olfactory and neuromodulatory signals can reverse visual object avoidance to approach in Drosophila by modulating visual pathways.