*PhD in Computational and Mathematical Biology (EPSRC funded, 3.5 yrs, Sept 2019 start) [re-advertised]* General information & apply (funding is primarily targeted at UK students): http://www.exeter.ac.uk/studying/funding/award/index.php?id=3386 Project description: http://www.exeter.ac.uk/codebox/phdprojects/Tabak-EPSRC-DTP-Project.pdf <http://www.exeter.ac.uk/codebox/phdprojects/Rankin-EPSRC-DTP-Project.pdf> *Closing date 13 May 2019. Interviews will be held at the University of Exeter Streatham Campus during the week of 10 June 2019.* Neurons and endocrine cells generate pulses of electrical activity. The patterns of electrical activity vary from cell-to-cell, and characteristics of these patterns (pulse duration, amplitude, timing) are critical to the function performed by the cell. For instance, electric pulse duration and frequency determines the quantity of hormone released by endocrine cells. Electrical currents are produced by ion channels, which act as non-linear electric conductances. The interactions between these non-linear conductances generate the complex patterns of electrical activity. Thus, the exact activity pattern of a given cell depends in a complex way on the exact distribution, or combination of weightings, of the different ion channels. To understand how different combinations of ion channel weightings result in a given pattern of electrical activity, we use mathematical models based on nonlinear differential equations that describe how the conductances vary with the cell electric potential and how they in turn change this potential. The problem with these models is that a lot of parameters are unknown, including the weightings of each channel. Electrophysiologists can record the electrical activity patterns from individual cells, but they cannot measure all the channel weightings from one cell. The objective of this funded PhD project is to extract information about the weightings from the measured electrical activity. The successful candidate will generate a database of channel weighting combinations and compute the resulting electrical activity pattern associated with each combination. They will then apply machine learning and/or topological data analysis to this database. This will allow them to deduce the relationships between weightings across models that generate electrical activity patterns selected from a subset of the database. By identifying the relationships between channel weightings across a highly heterogeneous population we can pinpoint the rules that regulate electrical activity across heterogeneous populations of cells. These methods will then be applied to existing datasets of electrophysiological recordings from electrically active cells. This project provides a unique opportunity to develop experience in machine learning and to receive training in mathematical modelling of neurons. This work will be done in close collaboration with experimentalists using cutting-edge methods that incorporate modelling and electrical recordings together. The student will therefore be exposed to multidisciplinary teamwork. If they so desire, they will also have the opportunity to learn electrophysiology and generate their own dataset of experimental electrical recordings. Candidates with quantitative backgrounds (mathematics, physics, engineering, computer science) are encouraged to apply to this 3.5 year PhD Scholarship. Programming experience, knowledge of dynamical systems theory and experience in biological modelling are a plus. The successful candidate will also be expected to travel to conferences to present their work. *For further information, please contact Dr Joel Tabak J.Tabak@exeter.ac.uk <J.Tabak@exeter.ac.uk> and Dr James Rankin j.a.rankin@exeter.ac.uk <j.a.rankin@exeter.ac.uk>* http://emps.exeter.ac.uk/mathematics/staff/jar226