116964 research outputs found
Sort by
Engineering of human enzymes for amino acid depletion therapy in cancer
Get PDFCancer cells undergo metabolic alterations to sustain their abnormal growth and proliferation. Many cancer cells are now characterized as ‘addicted’ or ‘auxotrophic’ to certain amino acids due to their increased metabolic demand. Serine is the second most-consumed amino acid by cancer cells after glutamine due to its primary contributions to one-carbon metabolism by donating one-carbon units utilized in the purine and thymidine biosynthesis. Serine is also involved in many other anabolic reactions, including biosynthesis of glutathione, phospholipid head groups, and sphingosine. Dietary serine restriction has been shown to suppress tumor growth in several cancer models. Suppression in tumor growth by serine restriction has been associated to result from reduced glutathione, purine, and sphingolipid biosynthesis by cancer cells. Systemic application of therapeutic enzymes is a superior means of depleting key amino acids in comparison to dietary restriction as 1) the depth of depletion, i.e., reduction in systemic concentration of the amino acid can be far more significant, 2) can be controlled/titrated, and 3) circumvents significant issues with patient non-compliance to dietary regimens. L-asparaginase is the most well-known and only FDA-approved enzyme for the treatment of acute lymphoblastic leukemia, thanks to the high demand for L-asparagine in leukemia cells. This success led to explorations of many other enzyme-mediated amino acid depletion therapies, including L-methionase, L-cyst(e)inase, and L-arginase. As the growth of numerous tumors depends upon external sources of serine, a catalytically relevant human enzyme that degrades L-serine would provide a novel therapeutic agent for treating serine-dependent cancers. Human serine dehydratase (hSDH) is one such enzyme that degrades serine to non-toxic products, however, its pharmacodynamic properties are not sufficient for clinical applications. The following chapters describe the engineering of hSDH variants with higher catalytic efficiencies, longer serum half-lives, and higher expression yields as candidate therapeutics for systemic serine depletion. We also show that engineered hSDHs decrease tumor growth in breast cancer models by reducing available serine levels in serum and tumor tissues. This work also includes studies aimed at the pharmacological optimization of another human therapeutic enzyme candidate.Biochemistr
Shaping National Security Decisions: An Insider’s View of The Insiders’ Game
Get PDFElizabeth Saunders’s The Insiders’ Game offers a rich perspective regarding how legislators, military leaders, and high-ranking civilian officials shape national security decision-making.LBJ School of Public Affair
Learning adaptive dynamics models for robust contact-rich manipulation
Get PDFAccurately pushing an object with unknown physical properties in a sim-to-real contact-rich manipulation task is a largely unsolved problem in robotics. This work proposes a method that utilizes temporal difference model predictive control (TD-MPC2) and an adaptation module to reduce the sim-to-real gap for these tasks. Rapid Motor Adaptation is used as a baseline to illustrate the proposed method's effectiveness. The TD-MPC2 algorithm is trained with an encoded representation of privileged information in an environment with randomized mass and friction conditions. The adaptation module is then trained using a history of observations and actions to match the encoded representation created from the TD-MPC2 algorithm. The algorithm presented in this paper is able to successfully complete a simulated box pushing task 21.80% of the time with a large range of friction and mass. However, this success rate is lower than TD-MPC2 trained without privileged information (51.41%), demonstrating the need for further research.Mechanical Engineerin
Preparation for bias, racial socialization competency, and Latinx youth mental health
Get PDFThe stress associated with ethnic-racial discrimination contributes to negative mental health outcomes in Latinx youth, including greater internalizing (Lorenzo-Blanco & Unger, 2015) and externalizing symptoms (Ponting et al., 2018). Parental preparation for bias messages theoretically buffer against the damaging impact of discrimination (Kiang et al., 2019). However, research has documented a negative association between preparation for bias and mental health among Latinx youth (Liu & Lau, 2013; Salcido & Stein, 2023). Parent racial socialization (RS) competency has been proposed to be another important component of racial-ethnic socialization (RES) that may shape the impact of preparation for bias on youth outcomes. Racial socialization competency is comprised of four components that influence RES message delivery: parental RS confidence, RS skill, and RS stress (general stress, call to action stress) (Anderson et al., 2020). The present study examined whether RS competency (confidence, skill, stress) moderated the relation between parental preparation for bias and youth mental health symptoms. Data from 210 Latinx parents (M[subscript age] = 42.63, SD = 8.68, 56% female) of youth ages 10 to 18 (M[subscript age] = 14.37, SD = 2.50, 46% female) were analyzed for this study. The majority of parents (71.4%) were born in the United States and spoke English at home (76.2%). Main effects revealed that RS confidence was negatively associated with parent-reported youth mental health symptoms, but did not moderate the association between preparation for bias and symptoms. There were no significant effects for RS skill. However, as predicted, RS stress (general stress, call to action stress) moderated the relation between preparation for bias and parent-reported youth mental health. Overall, these findings highlight RS parental competency as a promising intervention target for Latinx parents who aim to support their youth when facing racial and ethnic discrimination. In particular, they underscore the importance of supporting parents with high stress around RS conversations and generally supporting RS confidence in parents to improve Latinx youth mental health.Psycholog
The mechanics and chemistry of 2D material bubbles
Get PDFWhen two-dimensional (2D) materials are transferred onto a supporting substrate, matter trapped at the interface can spontaneously form bubbles with nanometer-scale dimensions. These bubbles are explored for novel nanoscale applications, including to strain-engineer 2D materials, and to explore chemical reactions under nanoscale confinement. However, strategies to deterministically control the properties and behavior of these bubbles, such as the strain level in the 2D material, or the chemical content of the trapped matter, remain unclear. This is due to a lack of fundamental understanding of the bubble composition, the adhesive behavior of the 2D material-substrate interfaces, and the nanoscale mechanical behavior of 2D materials, and how these properties affect each other. This dissertation aims at unveiling these unknowns. We first investigate the mechanics of these spontaneously formed bubbles by measuring the aspect ratios of 2D material bubbles, calculated as the bubble height divided by radius, for various interfaces. Bubbles for a given 2D material-substrate pair are shown to have self-similar aspect ratios independent of the bubble volume. We collaboratively developed a simple scaling law and rigorous theoretical model for liquid-filled 2D material bubbles, which predicts that the interfacial work of adhesion is related to the fourth power of the aspect ratio of the bubble and depends on the surface tension of the trapped liquid. Since our model highlights the importance of the surface energy of the confined liquid in controlling the bubble pressure and strain, we next attempt to analyze and control the chemical composition of 2D material bubbles. HOPG/MoS2 bubbles are fabricated in environments with varying relative humidities. The bubbles that form after layering are analyzed using a time-of-flight secondary ion mass spectrometer (ToF-SIMS). Secondary ion mapping shows that the bubbles contain water, hydrocarbons, and PDMS residues. We employ a quantitative analysis of the secondary ion yield within each bubble to compare the relative concentrations of key chemicals. Furthermore, we find that the morphology of the multicomponent chemical mixture within the bubble depends on the ambient humidity under which the bubbles are formed. The work presented here provides new insights into the understanding and control of the mechanics and chemistry of spontaneously formed 2D material bubbles.Materials Science and Engineerin
Behavior of drilled shaft footings under pure compression loading and proposal of 3D strut-and-tie method
Get PDFStrut-and-tie method was intended to provide more realistic strength estimation and detailing requirements for deep structural members such as deep beams, bent caps, and drilled shaft footings, also commonly referred to as pile caps. Two-dimensional strut-and-tie model was well refined and adopted in current AASHTO LRFD Bridge Design Specification (2020). However, design methodology of three-dimensional strut-and-tie method in drilled shaft footings supported by a grid of drilled shafts has been significantly slowly adopted, which results from insufficient code provisions and guidance, difficult implementation to define nodal geometry, and highly conservative results. Furthermore, detailing design of drilled-shaft footing has difference between states, districts, and even municipality basis. Most previous research projects tested considerably small-scale specimens compared with pile caps in common construction project and specimens have different details from current practice. In this study, comprehensive shear-critical loading tests of pile cap specimens were carried out with a variety of parameters-strut inclination, shaft size, face reinforcement ratio, and depth of pile caps. A series of large-scale specimens which are approximately half-scale compared to pile caps constructed in Texas were tested under pure compression at the column resulting in uniformly compression in drilled shafts. The results of experimental research were investigated by each design variables to evaluate mechanical behavior. The numerical analysis with validated finite element models examined a variety of cases of aforementioned parameters and additional design parameters. The guideline and examples of pile caps supported by shafts in uniform compression are developed based on the investigation of a series of tests and numerical parametric study.Civil, Architectural, and Environmental Engineerin
Moving toward more efficient polymer drag reduction
Get PDFThis dissertation presents a comprehensive investigation into the targeted delivery of polymer additives for skin-friction drag reduction in turbulent wall-bounded flows. Traditional approaches to polymer-based drag reduction rely on uniform injection, which often leads to inefficient material usage and uncertain performance. This work explores an alternative strategy: using inertial particles as carriers to deliver polymers selectively to flow regions that are dynamically relevant, such as coherent vortical structures.
The study begins by characterizing how bubbles interact with a simple two- dimensional vortex. With this simplified problem, bubble entrapment can be characterized by only two dimensionless parameters. Direct numerical simulations of turbulent channel flow reveal that particle entrapment within vortices is governed by the lifetime and evolution of turbulent vortical structures. Frozen and randomized velocity fields further isolate various time- and space-dependent influences and establish a strong connection between particle entrapment and coherent structures.
Building on this understanding, the dissertation examines how targeted polymer injection can disrupt vortex coherence using minimal material. Controlled simulations involving a Lamb–Oseen vortex and a vortex ring demonstrate that polymers placed near high-strain regions are significantly more effective in weakening coherent structures than uniformly distributed polymers. These findings are then extended to fully turbulent channel flow using viscoelastic simulations with the FENE-P model. Smart injection strategies based on particle-informed flow diagnostics lead to improved drag reduction per unit of polymer, compared to traditional injection schemes.
Overall, this work introduces and validates a flow-informed framework for polymer drag reduction, bridging fundamental fluid–particle dynamics with practical strategies for flow control. The framework itself offers guidance for flow control strategies, and the results suggest that intelligent, localized delivery of polymers informed by real-time flow features can reduce turbulence more efficiently and sustainably in wall-bounded turbulent flows than uniformly distributed polymers or non-targeting approaches.Aerospace Engineerin
Fungi in concealed and visible spaces : controlling variables for fungal growth and emissions
Get PDFOur homes and businesses are where we spend 90% or more of our time. So the fungal exposure we have in these places is important to understand and control. Fungal exposure has been mostly associated with negative health outcomes including asthma, upper respiratory inflammation, watering eyes, headache, and other negative symptoms. The culprits for these reactions are the mycotoxins found in fungal cells and spores that are produced to fight off other fungi and microbes as well as microbial volatile organic compounds (MVOCs) produced by both primary and secondary metabolisms of fungi. Decreasing these building occupant exposures involved investigating fungi from two different angles, light and water. The first was the effect of light on fungal emissions. Light has been studied for its ability to limit fungal growth which will limit spores and shed cells but the tools for measuring MVOC emissions in real time have not been available until very recently. The common household fungi investigated (Alternaria alternata, Epicoccum nigrum, Cladosporium cladosporioides, and Rhodotorula mucilaginosa) increased emissions when exposed to light while growing on house dust at common household humidity (50% RH). The fungi also emitted C₅H₈, most likely isoprene (10-20% increase in μg/hr in light), for which specialized sensors exist that are much less expensive and more portable than the Vocus PTR-Tof-MS used for these experiments. The second study was investigating conditions that may result in liquid water, which supports fungal growth, in one of the most sensitive parts of our homes. It examines conditions that lead to condensation events in the ductwork of our HVAC Systems. Forced air systems are one of the least often cleaned spaces in our homes and the most important system for spreading clean air throughout the building. Control of these systems using fan over run controls or constant on fan settings results in warm moistened air passing through cooler ductwork which causes condensation (up to 0.032 g/m2 per cycle) in a place shown to have a buildup of dust and fungal spores to grow on the dust. The results were modeled to give researchers ways to analyze why this phenomenon happens.Civil, Architectural, and Environmental Engineerin
Biomarkers predicting outcomes for melanoma patients receiving systemic treatment
Get PDFMelanoma, the deadliest form of skin cancer, is the fifth leading cancer diagnosis in the United States (U.S.). The management and outcomes for patients with stage IV metastatic melanoma have been dramatically improved with the approval of 14 targeted therapies and immune checkpoint blockade (ICB) regimens since 2011. Many of these have also gained FDA approval as adjuvant therapies for patients with surgically resected disease. In other cancers with surgically resectable disease, the FDA has approved the use of neoadjuvant treatment or pre-operative systemic treatment given the improved clinical outcomes compared to adjuvant treatment alone. Aggregated data has established pathologic complete response (pCR) as a surrogate endpoint for favorable long-term clinical outcomes. In melanoma, multiple studies have now reported benefit for patients treated with neoadjuvant therapy. Furthermore, the neoadjuvant platform has provided critical insight regarding biomarkers that predict response and resistance. Melanoma is the optimal test environment for novel ICB developments and many successes within the field have also been translated to other solid tumors. The development of a deeper understanding on neoadjuvant treatment approaches in melanoma can provide insights regarding biomarkers to predict response and resistance to new therapies. This may ultimately help accelerate the ability to provide personalized care for patients and accelerate future drug development strategies not only for patients with melanoma, but also other cancers. This dissertation advances the biomarker insights associated with outcomes to neoadjuvant systemic treatment for patients with melanoma. Given that ICB treatment in melanoma has paved the way for further development in other cancers, insights gained from this proposal are potentially translatable to other cancers as well.Pharmaceutical Science
On traffic state estimation and control in the world of connected vehicles
Get PDFThe increasing number of vehicles on existing roadway networks has put immense pressure on transportation systems across the globe. There is an increasing need for efficient control techniques that require minimal resources in terms of infrastructure, for example, variable speed limit (VSL) control, ramp metering (RM) control, route control, and traffic signal control among others. Traffic control strategies depend heavily on the availability of traffic data within the controlled network region to operate effectively. Unfortunately, the deployment and maintenance of traffic sensors to collect such data at a large scale can be financially infeasible. This necessitates the development of methods for accurately estimating traffic density on roadways with limited access to data. This work addresses the problem of traffic state estimation (TSE) in the presence of heterogeneous sensors which include both conventional fixed sensors as well as moving sensors such as Connected Vehicles (CVs). A state-space formulation is presented for the realistic second-order Aw-Rascle-Zhang (ARZ) model considering junctions which is important to model real highways with ramps. Moving Horizon Estimation (MHE) is implemented for TSE and compared with existing approaches with regard to accuracy and computational tractability. The impact of sensor noise and various strategies for querying CV data on the estimation performance is also considered. While there is a bulk of literature on traffic control techniques, there is still a need for scalable methods of controlling traffic applicable on a network level. This work further presents an analytical approach to model the problem of VSL control on road networks described by the Lighthill-Whitham-Richards (LWR) partial differential equation (PDE). The approach is based on the Lax-Hopf solution framework and is modeled into an optimization problem. The resulting problem, which is non-linear in the decision variables, is transformed into a mixed integer linear program (MILP). A numerical example is presented to show the effectiveness of the approach, including its application to a real-world highway network with multiple ramp connections. The method is compared to a classical Link Transmission Model (LTM) formulation of the VSL control problem. The combination of boundary flow control such as with RM and VSL control is a widely used strategy for highway corridors to maintain safety and throughput when the capacity downstream is dropped. However, most proposed models assume fixed and known demand, which is not always true in reality. Following the aforementioned formulation, a two-stage stochastic model is proposed to tackle the uncertainties in demand. The presented model assumes that the demand follows a discrete distribution and the speed limit can be chosen from a predefined set with reasonable values. The control is implemented in a rolling horizon framework. A case study is presented showing that the proposed model outperforms three deterministic models, which use the minimum, mean, and maximum values of the demand as model input. While ensuring the maximum throughput, the proposed model is demonstrated to be able to reduce the fluctuation of entry flow without significantly blocking upstream traffic. Increased ride-hailing and delivery operations in cities have led to a rise in curbside stoppages contributing to traffic congestion. This work additionally proposes an optimal controller for real-time curbside management using surrogate modeling. A nonlinear optimization problem is formulated to minimize the congestion on roadway segments caused by vehicles stopping on the segment and implemented in a model predictive control (MPC) framework. A hybrid simulation approach used to model stopping vehicles (moving bottlenecks) in the traffic stream is adopted to reproduce the output of the optimization problem, and due to its non-linearity is coupled with a meta-heuristic. Since repeated simulations are time-expensive, a surrogate model is employed to replace the simulation within the meta-heuristic algorithm. Several modeling techniques are compared based on their accuracy in reproducing solutions to the problem and computational tractability for real-time control under different configurations of simulation parameters. Finally, an application-based numerical study is presented to show the potential of the proposed approach in reducing the negative impacts of stopping vehicles and favorable computational properties. The ability to relay messages to vehicles in real-time to regulate their position and speed is further explored to improve network performance against congestion. This work investigates traffic control via connected and automated vehicles (CAVs) using novel controllers derived from the linear-quadratic regulator (LQR) theory. CAV-platoons are modeled as moving bottlenecks impacting the surrounding traffic with their speeds as control inputs. An iterative controller algorithm based on the LQR theory is proposed along with a variant that allows for penalizing abrupt changes in control speeds. The controllers use an extended cell transmission model (CTM) which considers the capacity drop phenomenon for a realistic representation of traffic in congestion. The impact of various parameters of the proposed controller on the control performance is analyzed. The effectiveness of the proposed traffic control algorithms is tested and compared with existing proportional-integral (PI) and MPC controllers from the literature. A case study using the TransModeler traffic microsimulation software is conducted to test the usability of the proposed controller as well as existing controllers in a realistic setting and derive qualitative insights.Civil, Architectural, and Environmental Engineerin