A systematic understanding of how genes control neuronal activity would give insights into how gene expression dictates the function of specific neuronal subtypes, and pinpoint therapeutic targets for neurological diseases. To enable scalable characterization of the role of specific genes in neuronal activity, we developed an arrayed CRISPR screening platform in human iPSC-derived neurons and a machine learning-based model, Plexus, to analyze the resulting datasets. Using this framework, we identified potential genetic modifiers of aberrant neuronal activity in frontotemporal dementia.
This work was led by Parker Grosjean, a graduate student in the Kampmann lab. It was a collaboration with the Laboratory for Genomics Research (LGR), GSK, and Adam Yala.
See here for the publication:
Parker Grosjean, Kaivalya Shevade, Cuong Nguyen, Sarah Ancheta, Karl Mader, Ivan Franco, Seok-Jin Heo, Greyson Lewis, Dehua Zhao, Bhairavi Tolani, Steven Boggess, Angelique Di Domenico, Erik Ullian, Shawn Shafer, Adam Litterman, Laralynne Przybyla, Michael J. Keiser, Jamie Ifkovits, Adam Yala & Martin Kampmann (2025) Network-aware self-supervised learning enables high-content phenotypic screening for genetic modifiers of neuronal activity dynamics. Nature Machine Intelligence 7, 2009–2025.
The preprint is available here: https://www.biorxiv.org/content/10.1101/2025.02.04.636489v2