UMD Theses and Dissertations
Permanent URI for this collectionhttp://hdl.handle.net/1903/3
New submissions to the thesis/dissertation collections are added automatically as they are received from the Graduate School. Currently, the Graduate School deposits all theses and dissertations from a given semester after the official graduation date. This means that there may be up to a 4 month delay in the appearance of a given thesis/dissertation in DRUM.
More information is available at Theses and Dissertations at University of Maryland Libraries.
Browse
Search Results
Item TOWARD ENSEMBLE-BASED DRUG DISCOVERY THOUGH ENHANCED SAMPLING(2023) Smith, Zachary; Tiwary, Pratyush; Biophysics (BIPH); Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Quantitatively assessing protein conformational dynamics and ligand dissociation are two problems of critical importance for computer-aided drug discovery. Both of these problems involve larger shifts in the protein conformation than are ordinarily considered in drug discovery efforts. Even though it is well known that proteins are best described as a dynamic ensemble of states, actually acquiring a representative ensemble, especially one with probabilities attached to states, has remained an elusive problem. Molecular dynamics can in theory capture the full ensemble with a long enough simulation but it would take millions of years to simulate the timescale needed to study drug binding or unbinding. Given this timescale problem, it is necessary to develop software solutions to accelerate the sampling of these important rare events. A number of enhanced sampling methods such as metadynamics have arisen to deal with this problem but the methods that are able to attain the fastest speedup also require a low-dimensional description of the system's dynamics. In this thesis, I will develop methods to describe protein dynamics with low-dimensional functions that can be used with enhanced sampling and apply these methods in an enhanced sampling pipeline. The methods developed will both perform variable selection finding a small set of descriptors for the protein dynamics and perform manifold learning to find a low-dimensional representation of the dynamics using this set of descriptions. This pipeline will be used to tackle both problems of conformational dynamics and ligand dissociation in a relatively automated manner. I will then describe how solving these problems in a high throughput manner could impact structure-based drug design efforts, and the work remaining to attain that goal.Item A BIG-DATA-DRIVEN FRAMEWORK FOR SPATIOTEMPORAL TRAVEL DEMAND ESTIMATION AND PREDICTION(2023) Hu, Songhua; Schonfeld, Paul; Civil Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Traditional travel demand models heavily rely on travel surveys, simplify future demand forecasting, and show low sensitivity in response to spatiotemporal dynamics. This study proposes a deep-learning-driven framework based on mobile device location data (MDLD) for estimating and predicting large-scale travel demand at both individual and aggregated levels. This study first introduces how raw MDLD should be parsed to distill trip rosters and estimate population flow. Based on derived information, this study reexamines relations between average population flow and its determinants such as built environment, socioeconomics, and demographics, via a set of explainable machine learning (EML) models. Different interpretation approaches are employed and compared to understand nonlinear and interactive relations learned by EML models. Next, this study proposes a Multi-graph Multi-head Adaptive Temporal Graph Convolutional Network (Multi-ATGCN), a general deep learning framework that fuses multi-view spatial structures, multi-head temporal patterns, and various external effects, for multi-step citywide population flow forecasting. Multi-ATGCN is designed to comprehensively address challenges such as complex spatial dependency, diverse temporal patterns, and heterogeneous external effects in spatiotemporal population flow forecasting. Last, at an individual level, this study proposes a Hierarchical Activity-based Framework (HABF) for simultaneously predicting the activity, departure time, and location of the origin and destination of the next trip, incorporating both internal (individual characteristics) and external (calendar, point-of-interests (POIs)) information. For each individual, HABF first predicts activities via an Interpretable Hierarchical Transformer (IHTF). IHTF can efficiently handle big data benefiting from its transformer-based design to avoid recursion. Meanwhile, loss functions used in semantic segmentation are introduced into IHTF to address imbalanced distributions of activity types. Then, a local plus global probabilistic generator is designed to generate locations based on predicted activities and historical places, allowing individuals to visit new or historically-sparse places. Analyses are performed on several real-world datasets to demonstrate the model's capability in forecasting large-scale high-resolution human mobility in a timely and credible manner. Altogether, this study provides sound evidence, practically and theoretically, of the feasibility and reliability of realizing data-driven travel demand estimation and prediction at different spatiotemporal resolutions and scales.