News from Mar 18, 2014
Our paper "Complex RNA Folding Kinetics Revealed by Single-Molecule FRET and Hidden Markov Models" has appeared online in JACS. (http://pubs.acs.org/doi/abs/10.1021/ja4098719)
We use a hidden-Markov-model (HMM) analysis to construct a model of the Mg2+-dependent folding dynamics of the ribozyme Diels-Alderase from single-molecule FRET data. The model reveals an unprecedented amount of detail with multiple conformational states which are kinetically linked on a hierarchy of timescales ranging from 10 ms to seconds. In particular, five states were found independently and consistently in two different FRET constructs and at three different Mg2+ concentrations.
The HMM procedure used in this publication addresses some of the most important obstacles in the analysis of experimental time-series with HMM algorithms. In particular, we suggest an approach to choose the number of hidden states such that the model is as detailed as possible but still passes a series of validation tests. The hidden state can be aggregated using kinetic cluster algorithms. We also suggest ways to deal with experimental artifacts, such as background noise, acceptor blinking and spectral cross-talk.