Advanced fitting procedure for AFM single-molecule data

Atomic Force Microscopy (AFM) can be used in single-molecule binding studies to measure parameters such as off-rates and binding distances. However, interpretation of the obtained bond rupture data is marred by the need to use relatively simple models to describe a potentially very complex unbinding process. The unbinding process is typically not describable by a fixed 1-dimensional barrier height of a single bond but is characterized by a spectrum of barrier heights. Moreover, even if data is filtered to exclude clear multiple rupture events, situations where multiple bonds break simultaneously, thus appearing as single rupture events, further complicate the picture. We present a new data analysis method, using least likelihood fits, that takes some of these effects into account. The method is tested on artificial data generated by Monte Carlo simulations and then applied to measure molecular bond rupture data, using the avidin-biotin system as a model system.