Morrissey Group: Quantitative biology of cell fate and tissue dynamics
The group develops mathematical and statistical modelling approaches to study the dynamics and fate choices of stem cells within tissues.
Mammalian tissues are highly dynamic systems, with tissue specific stem cells constantly making stochastic fate decisions, while committed cells migrate to specific locations to carry out the function of the organ. Despite the large number of stochastic events occurring simultaneously in the tissue, this complex process is tightly regulated, maintaining balanced cell numbers and precise cell compartment organisation. The group is focused on using quantitative approaches to characterise such dynamics, determine how they are altered in disease and to use this information to understand how cell fate decisions are made. We work in close collaboration with experimental research groups working for instance with epithelial tissues or blood.
In one example of this modelling approach we used a stochastic model of mutation acquisition and stem cell dynamics (See figure. Model left and measured data right). The mathematical model was able to explain the data patterns measured in intestinal tissue from a mouse that expresses a reporter protein activated by DNA mutations (Kozar, Morrissey et al Cell Stem Cell 2013). Using the model we were able to show that the number of stem cells contributing to the intestinal stem cell pool is considerably lower than previously thought. Additionally this quantitative description gave us a baseline to investigate what effect key oncogenic mutations have on stem cell dynamics (Vermeulen, Morrissey et al Science 2013).
In a more recent example we are working on haematopoiesis and the sequence of cell fate choices that lead blood stem cells to become, for instance, red bloods cells.
Our main focus is to understand cell fate using mathematical modelling. This often requires the development of computational tools to derive information from imaging and high throughput data. For example, our lab has developed automatic image segmentation methods and Bayesian inference models for high-throughput assays.