Automated lifespan measurements using machine learning to study aging in the model organism C. elegans
Veerle Rottiers, Professor
Nutritional Sciences and Toxicology
Applications for Fall 2024 are closed for this project.
Aging affects all organisms. Understanding the conserved mechanisms of aging could lead to new strategies for the prevention and treatment of age-associated disease. C. elegans, a small roundworm, is one of the most widely used model organisms to study aging. Worms live for about 3 weeks making it practical to study aging while having many of the same genes and organs as humans. Many of the key discoveries in aging research in the last 30 years were made using C. elegans including the discovery of lifespan extension by downregulation of insulin/growth factor signaling.
The goal of this project is to setup an image analysis pipeline using machine learning to be able to automatically measure the lifespan of C. elegans. We have recently acquired an automated system for monitoring the behavior of group-housed C. elegans throughout their lifespan called the “The C. elegans Observatory” (Kerr et al, 2022 https://pubmed.ncbi.nlm.nih.gov/36105851/). The Observatory can take images and movies of up to 384 6 cm agar dishes of C. elegans every 6 hours. High-speed image processing captures a range of behavioral metrics, including movement speed and stimulus-induced turning, and a data processing pipeline continuously computes summary statistics. One current limitation of the Observatory is that it does not report the measurements of the lifespan of C. elegans. The goal of this URAP project is to add such a functionality.
Role: Specifically, the student(s) will have to write a user-friendly program that takes the images of an agar plate containing dozens of C. elegans collected by the observatory every 6 hours for several weeks and return the number of dead and alive animals and their position on a plate at each timepoint. The specific tasks would involve everything from labeling images to testing various machine learning architectures to writing most of the code and relevant documentation.
Qualifications: Undergraduates seeking to apply should have previous experience with using machine learning for large-scale image analysis or be proficient in using machine learning for other applications. No background or experience in biology is required.
Day-to-day supervisor for this project: Denis Titov/Veerle Rottiers
Hours: to be negotiated
Related website: http://denistitovlab.org
Related website: http://denistitovlab.org