BACKGROUND:

​

Cytosolic calcium regulates multiple cellular processes, including gene expression, metabolism, enzyme activity, muscle contraction and hormone release. Maintaining the right cytosolic calcium concentration is therefore key for a cell’s survival and function. The cell regulates cytosolic calcium levels through electrical activity by opening and closing ion channels in the cell membrane. Electrically active cells (like cardiac myocytes or pituitary lactotrophs) must balance their function (pacemaking and hormone secretion respectively) with maintaining the desired gene expression. Gene expression in turn modifies the cell’s electrical activity, highlighting the complex interactions between different regulatory mechanisms in the cell.

​

How can one chemical signal successfully control so many aspects of a cell’s life? Recent modelling of gene expression suggests that a cell can change its concentration of ion channel proteins in the cell wall in response to changing cytosolic calcium targets. Myocardial cells contract in response to transient cytosolic calcium spikes. Experiments also suggest that hormone secretion can be driven by such transient spikes, and by increases in cytosolic calcium spiking frequency. It is possible that different cellular processes respond to various aspects of the cytosolic calcium signal.

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

 

METHODOLOGY:

​

My research is exploring how to model different regulatory mechanisms interacting, to propose different biological regulatory mechanisms that can be explored in experiments. A key challenge is to simulate large numbers of cells rapidly. Another is to accurately portray cellular processes that take hours (gene expression) and milliseconds (calcium spikes) in one model. I have therefore worked extensively with computational techniques to parallelise computations.

​

Further, I am investigating using techniques from Differential Geometry (a branch of mathematics) to implicitly model cell evolution, without explicitly simulating hours of gene expression. My goal is to increase modelling throughput by 1,000,000 times or more, going from 1,000 hours of cell simulation (which is typical today) to over 1 billion hours. This is necessary to truly understand regulatory mechanisms and propose further experimental studies for particular biological mechanisms.

​

​

​

​

​

​

​

​

​

​

​

​

​

 

 

RESULTS:

​

The main results thus far are around computational methods. I have derived a method to implicitly study cell evolution using Differential Geometry. By selecting the parameters of all models whose output fulfil certain dynamic criteria (a specific mean calcium level or transient size), it it possible to study how parameters evolve with respect to these criteria. This makes it possible to study how a model’s dynamics evolves in response to parameter changes. The method will drastically increase computational throughput.

​

Further, I have published a Python library (https://pypi.org/project/clode/) that lets users rapidly simulate cell action potentials and other ordinary differential equations. To lower barriers to entry, I created runtimes for Windows, Mac and Linux, letting researchers use the system with every major operating system. Python is also compatible with MatLab. The goal is to spread good science by designing easy-to-use tools and increase experiment replicability.

​

​

​