A1644
Title: Inference of transition time in steady-state variations in smFRET via a Wasserstein distance approach
Authors: Fengnan Gao - University College Dublin (Ireland) [presenting]
Abstract: Many biological molecules respond to external stimuli that can cause their conformational states to shift from one steady state to another. To study steady-state transitions in single-molecule fluorescence resonance energy transfer, a novel methodology is introduced, called WAVE (Wasserstein distance Analysis in steady-state Variations in smFRET) to detect and locate non-equilibrium transition positions in FRET trajectories. The method first utilizes a combined STaSI-HMM (stepwise transitions with state inference hidden Markov model) algorithm to convert the original FRET trajectories into discretized trajectories. Maximum Wasserstein distance analysis is then applied to differentiate the FRET state compositions of the fitting trajectories before and after the non-equilibrium transition. This methodology allows observing changes in experimental conditions in chromophore-tagged biomolecules or vice versa. Statistical properties, such as consistency and convergence rates, of the methodology are investigated under mild conditions.
