Scientists learn about the brain’s inner workings by studying what animals or people do, how they move, react, and make choices. Behaviour is complex, as animals and humans can move in countless different manners. Yet, neuroscience studies have heavily relied on constrained, simplistic behaviours, which are easier to analyse and quantify. Now recording and tracking technologies offer scientists access to movements of body parts of freely behaving animals at sub-second and millimetre scales. These detections of animal body parts often contain missing data that can hinder the analysis of both behaviour and neuroscience experiments.
A research team led by Professor Dr Katarzyna Bozek from the Center for Molecular Medicine Cologne (CMMC) and the CECAD Cluster of Excellence on Aging Research designed a method for recovering missing data in recordings of animal behaviour. The study “Deep Imputation for Skeleton Data (DISK) for Behavioral Science” was published in Nature Methods. It was made possible by collaborations with research labs at the Okinawa Institute of Science and Technology (OIST) in Japan, Vrije Universiteit Amsterdam, and the Salk Institute for Biological Studies in the U.S., which provided behaviour experimental data of mice and zebrafish.
The Deep Imputation for Skeleton Data method, short DISK, is based on a Transformer neural network architecture and operates across species, ranging from insects through fish to rodents. The method can recover and fill in missing data across all parts of the body and provide real-time estimations of the quality of the data recovery. “We aimed to make the tool usable by a broad community of behavioural researchers, without relying on prior knowledge of the species, the number of animals, or any specific information about the behavioural task,” said Dr France Rose, the first author of the study. Researchers using this new method can take full advantage of their behavioural experimental data, which are usually expensive and time-consuming to generate.
The researchers demonstrated that DISK can improve the robustness of scientific results of such experiments, e.g., the statistical power of comparing of step dynamics between two mice groups under different pharmacological treatments. In addition to imputation, DISK learns meaningful representations of movement sequences that capture motion features such as speed and direction, as well as the performed actions, such as walking or climbing. DISK will allow behavioural researchers to improve their data quality and analyses. “We also believe that the novel way of quantifying behaviour using Transformer neural networks will find broad applications in this field of research,” said Bozek.
This research was supported by the KI-Starter Grant from the Ministry of Culture and Science of North Rhine-Westphalia, awarded to Dr Rose, the Program for Female Junior Researchers in Artificial Intelligence of the previous German Federal administration’s Ministry of Education and Research, awarded to Professor Bozek, and additional support from the Japan Society for the Promotion of Science (JSPS). Dr. Rose is currently an Emmy Noether junior group leader at the University of Bonn.
Media Contact:
Professor Dr Katarzyna Bozek
Center for Molecular Medicine Cologne (CMMC)
+49 221 478 89529
k.bozek@uni-koeln.de
Press and Communications Team:
Dr Anna Euteneuer
+49 221 470 1700
a.euteneuer@verw.uni-koeln.de
Publication:
https://www.nature.com/articles/s41592-025-02893-y
Verantwortlich: Dr. Elisabeth Hoffmann – e.hoffmann@verw.uni-koeln.de