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


Bogetti, Anthony (2023) SHOOTING FOR THE MOON WITH WEIGHTED ENSEMBLE APPLICATIONS. Doctoral Dissertation, University of Pittsburgh. (Unpublished)

Download (24MB) | Preview


Rare events, which are infrequent, but relatively fast once they occur, are ubiquitous in biology. Conventional molecular dynamics (cMD) simulations can only sample rare events up to the microseconds timescale on typical resources because they spend most of their computing time sampling stable states. Path sampling strategies, which exploit the separation of timescales inherent in rare events, focus computing power on the transitions between stable states and are orders of magnitude more efficient than cMD. The weighted ensemble (WE) strategy is a particularly promising path sampling strategy that has not yet reached its full potential. In this dissertation, I describe various advances to the WE strategy that I have developed and demonstrate how those advances have allowed us to “shoot for the moon” by simulating larger systems on longer timescales. In Chapter 1 of this dissertation, I motivate the need for path sampling and discuss the features of WE that set it apart from other path sampling strategies. In Chapter 2, I describe various protein conformational switches and how path sampling strategies can rationally enhance switching kinetics by focusing sampling on the transient states of switches. This chapter highlights WE simulations of the SARS-CoV-2 protein, showcasing the ability of WE to generate complete pathways for systems up to a million atoms and processes as slow as the seconds timescale. In Chapter 3, I present advances to the open-source WESTPA software package (version 2.0) that were motivated by a recent SARS-CoV-2 “stress test.” In Chapter 4, I introduce a minimal, adaptive binning (MAB) scheme for WE simulations and showcase the MAB scheme using three systems of varying model resolution. In Chapter 5, I describe LPATH, a general, semi-automated tool for performing bottom-up clustering of simulated pathways into distinct classes using a text-string pattern matching algorithm commonly used in plagiarism detection. Together, the above chapters demonstrate the potential of WE path sampling to “shoot for the moon”, tackling larger systems and/or longer timescales.


Social Networking:
Share |


Item Type: University of Pittsburgh ETD
Status: Unpublished
CreatorsEmailPitt UsernameORCID
Bogetti, Anthonyatb43@pitt.eduatb430000-0003-0610-2879
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Committee ChairChong,
Committee MemberCoalson,
Committee MemberHutchison,
Committee MemberZuckerman,
Date: 5 September 2023
Date Type: Publication
Defense Date: 29 June 2023
Approval Date: 5 September 2023
Submission Date: 17 July 2023
Access Restriction: No restriction; Release the ETD for access worldwide immediately.
Number of Pages: 122
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: molecular dynamics, weighted ensemble, covid, cars-cov-2, spike protein, westpa, enhanced sampling, path sampling
Date Deposited: 05 Sep 2023 16:49
Last Modified: 05 Sep 2023 16:49


Monthly Views for the past 3 years

Plum Analytics

Actions (login required)

View Item View Item