I was awarded £800 flexible funding to cover registration on SYSMIC Module 1, an online course aimed at bioscientists. It provided training in computer simulation of biological systems using MATLAB and statistical analysis using R, as well as the mathematical background required for this. The course began in November 2017 and consisted of eleven 2-week sessions, followed by a longer mini-project which enabled integration of the skills learned for a more in-depth analysis of a modelling problem.

Before beginning this course I had no experience of using MATLAB or R, or of writing scripts in any software. I found that the course gave a good introduction to both pieces of software, and explained basic commands and the relevant syntax in an easily understandable way. In each session, the course provided information on the biological problem being tackled and the steps one would take to model it in a computer system. At many points, additional exercises were provided which enabled me to use what I had learned without blindly copying from the textbook, which helped a great deal in becoming more familiar with the software.

The first eight sessions focused on using MATLAB to model biological systems, before moving to R for the last three. Whilst I understand that the course (and final mini-project) were aimed at modelling biological systems, I would have appreciated an introduction to R earlier in the course and I felt that it was rather heavily skewed towards MATLAB. Although MATLAB has been very useful to me, I do feel that learning to write scripts in R would be a more generally useful skill – the high cost of MATLAB means that R is more widely used and accessible.

For the mini-project, I used MATLAB to replicate two models of neuronal firing in the mammalian brain (the FitzHugh-Nagumo and Rinzel models). These models use systems of coupled differential equations to simulate changes in membrane potential based on input current, and differ in whether or not they incorporate feedback into the system. I was able to replicate these models and use them to simulate regular- or burst-firing patterns in neurons, as well as generating phase-space plots and investigating the effect of introducing a random noise element to the system. Being able to apply multiple elements of the course within a larger project was very useful, as was the opportunity for critical discussion of these computer simulations and how they could be improved.

In general I enjoyed the course, finding it to be engaging and set at an appropriate level for someone not familiar with modelling. Early in 2018 my project workload had increased substantially and I found keeping up with the fortnightly assignment submission deadlines difficult, however the course was flexible about this and I managed to complete all assignments by the final deadline. Since completing the course my project has involved a large amount of data analysis and manipulation, and I have been able to use both R and MATLAB for tasks which have been invaluable to this (and might have been extremely difficult otherwise). To summarise, I have learned new methods in data analysis from this course which are already of use to me in my research, and I am glad to have had the opportunity.