Martin Lindén: Mean field theory and Bayesian inference in single molecule biophysics

Single molecule experiments opens new windows to molecular biology and biophysics, by allowing us to follow individual proteins at work in real time. However, instrumental artifacts and the inherent randomness of Brownian motion and low copy number chemistry often makes for noisy data that can be challenging to interpret. A very common problem is to analyze noisy time series with abrupt changes, reflecting for example binding events, or conformational changes in a protein complex. I will describe an approach to tackle such problems using hidden Markov models and a Bayesian version of mean field theory, and show results for single particle tracking of single proteins in live E coli cells. By modeling changes in diffusion constant (due to binding of small fluorescent proteins to large binding partners), we are able to extract interconversion rates between separate diffusive states, thus turning our setup into a non-invasive probe for intracellular kinetics. reference: Persson, F., Lindén, M., Unoson, C. & Elf, J. Extracting intracellular diffusive states and transition rates from single-molecule tracking data. Nature Methods 10, 265–269 (2013).