In a 2025 Science Advances publication titled “PseudoSorter: A self-supervised spike sorting approach applied to reveal Tau-induced reductions in neuronal activity,” a collaborative team led by Gabriele S. Kaminski Schierle introduces PseudoSorter, a cutting-edge spike sorting method that uses self-supervised machine learning to detect neuronal activity with higher precision.
The researchers applied this method to microelectrode array (MEA) recordings of hippocampal neurons exposed to subneuronal levels of extracellular Tau, a protein associated with Alzheimer’s disease. Their findings? Even minimal Tau exposure significantly reduced firing in specific neuron populations—effects that were undetectable using traditional analysis methods. To capture these nuanced electrophysiological signatures, the study used NeuroNexus-32-channel microelectrode arrays, enabling dense, high-resolution recordings critical for neuron-level classification.
This study not only enhances our understanding of Tau-related pathology but also showcases how advanced ML tools and NeuroNexus technology are redefining the frontier of neurodegenerative research.
