Development and Application of a Novel Enhanced Sampling Method and Bayesian Analysis for Characterizing Intrinsically Disordered Proteins

Intrinsically disordered proteins (IDPs) are a class of proteins with wide-rangingsignificance in signaling and disease that do not adopt a dominant folded structure asmonomers. Rather, the structures of IDPs in solution are best described as ensembles ofconformational states that may range from being fully random coil to partially ordered.This structural plasticity of IDPs is theorized to facilitate regulation of their interactionwith other species, as in signal transduction or aggregation of IDPs into ordered fibrils.Characterizing the structural ensembles of IDPs in the free, solvated state is key tounderstanding the mechanisms of these interactions, and correspondingly the role an IDPspecies plays in signaling or disease.The rapid interconversion between conformational states, however, complicates theexperimental study of IDPs because most experimental signals report highly averagedinformation. Computational modeling with validation through comparison to experimenthas therefore been a main approach to characterizing IDP structure and dynamics. Thefocus of my dissertation is on the development of new methods for computational study ofIDPs, facilitating better and less expensive de novo generation of IDP structural ensemblesand improving the metrics used to evaluate the degree of agreement between a simulatedensemble and a set of experimental data.Despite vast improvements in computational power and efficiency, moleculardynamics (MD) simulations of IDPs for generating conformational ensembles are stilllimited by the expense of calculations. In Chapter 2 I present the development of a newenhanced sampling method – temperature cool walking (TCW) – and comparison of itsperformance against a standard method – temperature replica exchange (TREx). The TCWmethod accelerates the rate of convergence to the equilibrium conformational ensemblewith increased sampling acceleration relative to TREx at greatly reduced computationalcost.The second major limitation in MD is the accuracy of the force field. Most classicalfixed charge force fields were parameterized using data from folded proteins, and havebeen thought to be biased to overly collapsed and structured conformations. This hasmotivated the development of IDP-tailored force fields that sample greater disorder, at thepotential expense of the ability to model stabilizing interactions between an IDP and itsbinding partners. In Chapter 3, I assess to what degree the shortcomings assigned tostandard force fields may be due to insufficient sampling by characterizing theperformance of standard and newly modified force fields on the Alzheimer’s peptideamyloid-β using both TREx and TCW. We find that with improved sampling, standard andmodified force fields produce similar structural ensembles, suggesting that both areappropriate for simulation of the disordered state. In Chapter 4 I present preliminaryresults building off of this work by characterizing the performance of a polarizable forcefield modeling a synthetic peptide that demonstrates complete loss of helical content withincreasing temperature. Inclusion of polarization effects has been thought to be key foraccurate modeling of such multicomponent systems, especially when there is a shift in theelectrostatic environment as is the case for the unfolding peptide. Our early results, whilelimited by current lack of convergence for tests using the polarizable force field andneeding further confirmation, match that expectation by finding early evidence of greaterresponse to temperature by the polarizable force field than fixed charge comparators.The last work presented here is in the development of new methods for calculatingthe degree of agreement between a simulated IDP ensemble and experimental data. Backcalculationof experimental data from structure can be very imprecise, motivating thedevelopment in Chapter 5 of scoring formalisms that account for variable uncertainties inboth back-calculation and experiment for diverse experimental data types. In summary, themethods described in this dissertation seek to improve computational study of IDPs byfacilitating better, less expensive generation of IDP ensembles and producing moreinformative metrics for evaluating their agreement with experiment