J. Kaestel-Hansen

This paper introduces DeepSPT, a deep learning framework that automates the analysis of sub-cellular diffusion to infer functional information, achieving high accuracy in identifying endosomal structures during viral infections.

SEMORE automates super-resolution data analysis by using machine learning to segment and quantify protein assemblies, enabling classification of aggregation pathways and nuclear pore complex geometry.

This paper uses single-molecule imaging to study DNA and nucleosome condensation by polyamines, HP1α, and Ki-67, revealing kinetic differences in condensation and dissolution.