Dresden 2026 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
BP: Fachverband Biologische Physik
BP 15: Computational Biophysics III
BP 15.9: Talk
Wednesday, March 11, 2026, 12:00–12:15, BAR/SCHÖ
Deep-Pose-Tracker: a unified model for behavioural analysis
of Caenorhabditis elegans — •Debasish Saha1, Shivam Chaudhary1, Dhyey Vyas2, Anindya Ghosh Roy2, and Rati Sharma1 — 1Indian Institute of Science Education and Research Bhopal, Bhopal, India — 2BRIC-National Brain Research Centre, Gurugram, India
The ability to respond to environmental stimuli by living organisms is essential for survival and adaptation. Locomotion and posture-based analyses of animals are commonly performed; however, manually performing these tasks is effort-intensive, time-consuming, and error-prone. Automation of this process is therefore crucial for accurate and fast detection. To this end, in this work, we report the development of Deep Pose Tracker (DPT), an end-to-end deep learning model to automate the study of posture dynamics and locomotion behaviour of C. elegans, a model organism useful to study neuroscience, genetics, drug design, etc. The DPT model enables automatic detection and tracking of these animals while measuring essential behavioural features like locomotion speed, orientation, forward or reverse locomotion, complex body bends as omega turns, and eigenworms (representing the overall posture dynamics in a low-dimensional space). Our DPT model can generate highly accurate data, with very high inference speed, while being user-friendly and robust to experimental variabilities. DPT, therefore, can be a valuable toolkit for researchers studying behaviour under different environmental stimuli.
Keywords: Animal behaviour; C. elegans; Pose detection; Tracking; Deep learning