Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

Efficient Computer Simulations of Protein-Peptide Binding Using Weighted Ensemble Sampling

Zwier, Matthew C (2013) Efficient Computer Simulations of Protein-Peptide Binding Using Weighted Ensemble Sampling. Doctoral Dissertation, University of Pittsburgh. (Unpublished)

Primary Text

Download (26MB) | Preview


Molecular dynamics simulations can, in principle, provide detailed views of protein-protein association processes. However, these processes frequently occur on timescales inaccessible on current computing resources. These are not particularly slow processes, but rather they are rare — fast but infrequent. The weighted ensemble (WE) sampling approach provides a way to exploit this separation of timescales and focus computing power efficiently on rare events. In this work, it is demonstrated that WE sampling can be used to efficiently obtain kinetic rate constants, pathways, and energy landscapes of molecular association processes. Chapter 1 of this dissertation further discusses the need for enhanced sampling techniques like the WE approach. In Chapter 2, WE sampling is used to study the kinetics of association of four model molecular recognition systems (methane/methane, Na+/Cl–, methane/benzene, and K+/18-crown-6 ether) using molecular dynamics (MD) simulations in explicit water. WE sampling reproduces straightforward “brute force” results while increasing the efficiency of sampling by up to three orders of magnitude. Importantly, the efficiency of WE simulation increases with increasing complexity of the systems under consideration. In Chapter 3, weighted ensemble Brownian dynamics (BD) simulations are used to explore the association between a 13-residue fragment of the p53 tumor suppressor and the MDM2 oncoprotein. The association rates obtained compare favorably with experiment. By directly comparing both flexible and pre-organized variants of p53, it is shown that the “fly-casting” effect, by which natively unstructured proteins may increase their association rates, is not significant in MDM2-p53 peptide binding. Including hydrodynamic interactions in the simulation model dramatically alters the association rate, indicating that the detailed motion of solvent may have substantial effects on the kinetics of protein-protein association. In Chapter 4, an all-atom molecular dynamics simulation of p53-MDM2 binding is described. We obtain an association rate that agrees with the experimental value. The free energy landscape of binding is “funnel-like”, downhill after the initial encounter between p53 and MDM2. Together, the studies described here establish that WE sampling is highly effective in simulating rare molecular association events.


Social Networking:
Share |


Item Type: University of Pittsburgh ETD
Status: Unpublished
CreatorsEmailPitt UsernameORCID
Zwier, Matthew
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Committee ChairChong, Lillian Tltchong@pitt.eduLTCHONG
Committee MemberCoalson, Rob Acoalson@pitt.eduCOALSON
Committee MemberJordon, Kenneth Djordan@pitt.eduJORDAN
Committee MemberZuckerman, Daniel Mddmmzz@pitt.eduDDMMZZ
Date: 30 September 2013
Date Type: Publication
Defense Date: 6 August 2013
Approval Date: 30 September 2013
Submission Date: 30 July 2013
Access Restriction: 5 year -- Restrict access to University of Pittsburgh for a period of 5 years.
Number of Pages: 121
Institution: University of Pittsburgh
Schools and Programs: Dietrich School of Arts and Sciences > Chemistry
Degree: PhD - Doctor of Philosophy
Thesis Type: Doctoral Dissertation
Refereed: Yes
Uncontrolled Keywords: computer simulation, molecular dynamics, weighted ensemble, protein binding, p53, MDM2, Brownian dynamics, kinetics
Date Deposited: 30 Sep 2013 22:30
Last Modified: 30 Sep 2018 05:15


Monthly Views for the past 3 years

Plum Analytics

Actions (login required)

View Item View Item