Using machine learning and live imaging tools to determine the dynamics of signals controlling stem cell fate decisions

Project ID: 2228cd1325 (You will need this ID for your application)

UCL Lead department: Division of Biosciences

Lead Supervisor: Marc Amoyel

Project Summary:

Stem cells have the unique ability to adopt different identities, either self-renewing to produce new stem cells, or differentiating to replenish tissues. How these decisions are made at an individual cell level is poorly understood. Previous approaches to determine how cells choose what fate to adopt have relied on genetic tools, but fail to capture the complexity of cell decision-making, both because they do not have temporal resolution and because they reduce complex decisions to single parameters. Here we will adapt new technologies in cell tracking and machine learning to identify parameters that predict cellular behaviour. We will use a combination of wet and computational lab techniques to image stem cells live in their endogenous environment, and determine how cell decisions are made. For example, this method can determine the answer to a question that has remained elusive to date: when does a stem cell lose plasticity and irreversibly commit to a differentiated fate? In addition, we will employ synthetic biology engineering methods to design and test tools to visualise cell-cell communication in real time. While such tools are starting to be developed, most are confined to cell culture or bacterial colonies, and fail to capture the complexity of a tissue, where homeostasis requires that different cells behave differently, to ensure that both stem cells and differentiated cells are produced in equal numbers. By combining these reporters with machine learning analysis, we will determine the engineering principles that underpin the maintenance of complex tissues.