Statistical Analysis of Single-Molecule AFM Force Spectroscopy Curves

We develop a statistical framework for the quantitative characterization and analysis of force-extension curves obtained from single-molecule force spectroscopy (SMFS) measurements. We apply this methodology to force-extension data obtained for elastinlike polypeptides (ELPs) with precisely engineered molecular architectures, where we demonstrate that our approach enables SMFS to be used to study hydrophobic hydration in intrinsically unstructured biomacromolecules. Our results obtained for ELPs suggest that hydrophobic hydration, rather than local backbone conformational entropy, is the key contributor to modulating the molecular elasticity of ELPs under changes in amino acid sequence. As with previous analysis, we parametrize SMFS curves using models from polymer statistical mechanics; however, we introduce several statistical innovations that dramatically improve the precision of the estimated parameters. Our approach (i) accounts for increased thermal noise in the data at low forces, (ii) provides confidence intervals for fitted polymer-theory parameters obtained from nonparametric bootstrapping, (iii) treats multiple curves simultaneously to reduce variability in the fitted parameters, and (iv) treats noise in both force and extension measurements simultaneously. A key advantage of our approach is the ability to quantify uncertainty in the fitted parameters, allowing comparison of parameters obtained under different experimental ∗Corresponding author: Scott C. Schmidler, Department of Statistical Science, Duke University, Durham, NC 27708-0251. Tel: (919) 684-8064; Fax: (919) 684-8594; Email: [email protected]