New tool to accelerate Alzheimer's research and drug discovery

A new machine learning method is giving scientists a more accurate way to analyse neural activity, offering fresh insights into diseases like Alzheimer’s and speeding up drug discovery.
The research, led by Marius Brockhoff and Jakob Träuble under Principal Investigator Gabriele Kaminski Schierle at the Department of Chemical Engineering and Biotechnology (CEB), is published today in Science Advances.
PseudoSorter is a new, highly accurate tool created to analyse the brain. It improves on existing techniques for sorting neuron signals, making it easier to detect subtle changes in brain activity. When tested on neurons from the hippocampus – an area of the brain crucial for memory – that were exposed to monomeric Tau, a protein linked to Alzheimer’s, it revealed a previously unrecognised drop in firing rates. This suggests Tau disrupts hippocampal function in ways that conventional methods might miss. Understanding these effects could help researchers to uncover how Alzheimer’s Disease affects the brain and, therefore, identify new treatment strategies.
First author Marius Brockhoff said: “Seeing that our research, combined with a range of experiments in collaboration with many talented scientists, led to new insights makes me very happy. The particular insight into the molecular mechanism underlying neurodegenerative diseases, while simultaneously providing a new analysis tool for further research, is something we are very proud to have achieved.”
The Molecular Neuroscience research group is a interdisciplinary group of scientists investigating molecules and mechanisms causing brain cells to die in different neurodegenerative diseases such as Alzheimer’s and Parkinson’s disease.
The Molecular Neuroscience research group is a interdisciplinary group of scientists investigating molecules and mechanisms causing brain cells to die in different neurodegenerative diseases such as Alzheimer’s and Parkinson’s disease.
Neurodegenerative diseases like Alzheimer’s are complex, and small but significant changes in neuron activity can be difficult to detect.
PseudoSorter makes it possible to identify these patterns with greater accuracy, giving scientists a more detailed picture of how diseases progress at the cellular level. Beyond Alzheimer’s, the tool could improve research into other brain disorders, such as Parkinson’s and motor neuron disease.
It could also speed up drug screening, helping researchers test how new treatments could affect neurons with greater precision. By improving efficiency, PseudoSorter could reduce the need for unnecessary experiments, making research more sustainable and minimising reliance on animal testing.
With its ability to handle large-scale data with high accuracy, PseudoSorter could become a valuable tool for neuroscience research, leading to better treatments and a deeper understanding of how the brain works.