Janne's research focuses on simulating the brain's neuronal wiring diagram using deep learning, recently earning him a first-author publication in Nature. In addition to his research, Janne also works to organize mental health initiatives for the community.
Janne is a doctoral candidate at the University of Tübingen and joined the IMPRS-IS research school in 2021. He works with Prof. Jakob Macke's Machine Learning for Science group and collaborates with Dr. Srinivas Turaga and other HHMI Janelia Research Campus researchers. He holds a Master's in Neuroengineering from TU Munich and a Bachelor's in Physics from the University of Göttingen.
His research focuses on simulating the brain using its neuronal wiring diagram (connectome) measured from electron microscopy. In his PhD, he developed a realistic simulation of the fly visual system using the connectome and deep learning. The resulting models accurately predict neural activity across the fly visual system at single-neuron resolution, validated against 26 experimental studies.
The team's work, recently published in Nature and titled "Connectome-constrained networks predict neural activity across the fly visual system," demonstrates how connectome measurements and machine learning can enable a "model-first" paradigm for neuroscientific discovery.
Now, he is guiding other students in Jakob's lab in extending this research while writing his PhD thesis. He is also improving model optimization techniques while maintaining the model's realism.
Beyond research, he enjoys running along the Neckar or in the Schönbuch, swimming, cooking, and music. Passionate about promoting mental health literacy in academia, he recently co-organized the Healthy Minds workshop and talk series.
Reach out through the contact on his website to connect over research or initiatives.
“How can we construct models that perform meaningful computations and are biologically realistic?” Janne presents his preliminary results at the Champalimaud Research Symposium: Dialogues on Neural and Machine Intelligence, Lisbon, Portugal, 2021.
More information