The goal of this project is to use machine learning (ML) to identify senescent cells—cells that have stopped dividing and contribute to aging and diseases like cancer—in large biological datasets, including data from gene expression (transcriptomics) and cell imaging. Traditional methods that rely on specific biological markers to find these cells are often limited in terms of scale and accuracy. In this project, we will test different ML techniques, such as supervised learning, unsupervised learning, and deep learning, to improve how we detect and classify these senescent cells. The main questions we aim to answer are: How can ML accurately identify these cells? Which data features are the best indicators of senescence? And what new biological information can we learn from these models?
The results will provide a scalable, data-driven framework to enhance our understanding of senescence in biological systems. This project lies at the exciting intersection of computational science and biology, offering the student the opportunity to contribute to both fields. The main supervisor has extensive experience in developing computational tools for understanding tissue heterogeneity, as demonstrated by key publications in Cell (2021), Cancer Cell (2021), and Nature Biotechnology (2023).
Prerequisites
- The student will be trained in understanding the data, and how to apply existing tools.
- The student needs to know some programming (for example Python and/or R) and must have a genuine interest in computational analyses.
- Contact:
- Main supervisor: Chloé B. Steen chloebs@uio.no (https://www.ous-research.no/cbs)