11:00-11:50 Gene expression dynamics in neural development
Dr Cerys Manning, MRC Career Development Fellow, Division of Developmental Biology and Medicine, University of Manchester
12:10-13:00 Stochastic simulation, analysis and inference for reaction networks
Neural development is a dynamic process with cell state transitions between neural progenitor cells and differentiated cell fates such as neurons, yet many of the current techniques to study it only provide a single timepoint snapshot. While we have a good understanding about general molecular determinants of neural cell fate, it is surprising how little we know about the dynamic gene expression changes during transitions between progenitor and differentiated states. It is vital that this knowledge gap is filled because these cell state transitions are critical for tissue development and when mis-regulated can lead to developmental disorders. Recent evidence shows that cell state transitions are not merely binary on-off switches in gene expression. In this talk I will discuss my work using live imaging of neural development. I will discuss how changes in the expression dynamics of the key cell fate determinant HES5 are concurrent with and important for cell fate transitions between neural progenitors and neurons. We will discuss changes in dynamics from noisy aperiodic to ultradian (short-period) oscillations, and how mathematical modelling of dynamics can inform on mechanism.
Dr Giorgos Minas, Lecturer in Mathematics and Statistics, School of Mathematics and Statistics, University of St Andrews
The continuous biotechnological advances, particularly over the last two decades, continue to provide larger and more informative datasets. They promise a more insightful understanding of biological processes and biomedical advances. Analogous mathematical and statistical advances are required to support these technological advances. A critical challenge to overcome is that biological data are often highly variable due to multiple sources of non-trivial variability affecting them. Stochastic models for reaction networks can describe biological processes incorporating their stochasticity. They can also describe epidemiological, ecological, and sociological processes. In this talk, I will introduce the main approaches for developing stochastic models of reaction networks. I will then describe a new approach for stochastic modelling that achieves a suitable balance between model accuracy, computational speed, and scalability to large systems. We will discuss methods for long-time stochastic simulation, analysis of parameter sensitivities, and statistical inference using time-series data. The method will be applied to large biological systems with oscillatory dynamics.