General conditions
While I spend most of my time teaching and coordinating courses, I am happy to supervise a couple of MBLS/Biology students at a time for their BSc thesis. Students interested in my field must be able to prove sufficient physics, math and programming background.
To give you the best supervision possible (and to ensure that your project is more interesting!), I am only open to part-time projects that span over 20 weeks/2 periods.
Please note that my projects are generally flexible and could be tailored to be more specific after a face-to-face meeting.
Possible topics to explore under my supervision
| Topic | Description | Ideal for students with… | |
|---|---|---|---|
| Computational Neuroscience | Any system that can be expressed as a first order differential equation and subject to Gaussian noise can be studied through the Fokker-Planck equation. This powerful tool from statistical physics predicts the evolution in time of the probability density of a certain system, proving useful in the study of firing rates and patterns. In this project you will apply this formalism to noise in neuronal models and compare the results of individual simulations with the statistical simulated prediction. | …a very strong math background, good Python skills and sufficient neurophysiology understanding. | |
Neuronal behaviour can be modelled through mathematical equations on several different levels: molecular, compartmental, cellular, synaptic and collective (in a network). These equations can then be simulated numerically to mimic real electrophysiological traces and analyze emergent dynamics. In this project you will consider one or more neuronal models and run some simulations to analyze firing patterns under varying conditions. | …a good math background, good Python skills and sufficient neurophysiology understanding. | ||
Ion channels and more in general membrane proteins can be represented mathematically by time-continuous finite-state Markov chains. Different models result in different molecular behaviours, especially when some parameters become voltage-, time- or light-dependent. In this project you will model the states of a membrane protein/ion channel and study the stochastic behaviour and/or the population evolution, either by numerical simulations or through analytical treatment. | …a good math background and good Python skills. | ||
| Nanophotonics | Metal nanoparticles find extraordinary applications in biology, especially when it comes to their coupling with fluorescent molecules (plasmonic enhancement). To be most effective, the properties of such particles such shape, size and material must be precisely tuned. | Most researchers in the field spend time and money into endless numerical simulations until they find the optimal parameters. However, an elegant rigorous analytical treatment already exist for ellipsoids (LINK HERE) and most importantly is open-source. In this project you will create a graphical user interface (GUI) that embeds the analytical model, and extend it for other morphologies based on the Bergman spectral decomposition. The outcome of this project might be made open-source and eventually even published in a paper. | …a good physics and math background, and very strong Python skills. Prior knowledge of 3D modelling (e.g. in Blender) or GUI design is a plus. |
Systematic validation of different combinations of material, size and morphology require numerical simulations. In this project you will methodically simulate the spectra of metal nanoparticles with finite-difference time-domain (FDTD) or other computational methods. These will be extremely useful to validate the analytical model explained above. | …a sufficient physics background, and sufficient programming skills. Prior knowledge of 3D modelling (e.g. in Blender) is a plus. | ||