137091 research outputs found
Sort by
Impact of Technological Uncertainty on Design Parameter Selection of the NASA Parallel Hybrid-Electric Propulsion EPFD Vehicle
No full textWhile electrified aircraft propulsion (EAP) technologies appear as a promising solution to enable more sustainable commercial aviation in the near term, their continued development within this time frame remains subject to technological barriers. As a result, the performance of future vehicles making use of these technologies, such as the EPFD vehicle, is inherently uncertain. Previous efforts have focused on the application of sensitivity analysis and uncertainty quantification methods to develop insights and assess the potential benefits that such hybrid-electric propulsion architectures may provide over conventional aircraft. These initial efforts consisted in propagating technological uncertainty on a previously fixed vehicle design, the nominal design point. In this paper, we show that this approach suffers from two main limitations: 1) performance projections are impacted by the choice of the nominal design point, and 2) design constraints are likely to be violated by the design combinations resulting from the uncertainty propagation process. To address these limitations, we propose two alternate approaches to uncertainty propagation that better account for the EPFD vehicle design problem, including the associated constraints. The first, design optimization under uncertainty (DO-U), relies on principles from robust design and reliability-based design to find a nominal design point that is robust to technological uncertainty. The second, uncertainty propagation on design optimization (UP-DO), does not seek a nominal design point: instead, it aims at quantifying the impact of technological uncertainty on the optimal vehicle designs, i.e. designs that were optimized to maximize performance given a known set of available technologies. It was found that both proposed approaches effectively address the previously identified limitations. If a nominal design point is required, then DO-U should be preferred. However, if the goal is solely to assess the benefits of EAP technologies, then uncertainty propagation on design optimization (UP-DO) better accounts for the full potential of these technologies, as it does not impose a potentially unnecessary requirement for robustness
Engineering Polymers of Intrinsic Microporosity for CO2/CH4 Separations
No full textThe conventional approach to designing and synthesizing polymeric materials for industrial applications has traditionally relied on experimentalist judgment of material structures and properties. After material selection, the synthesis of the polymer and subsequent performance testing involves extensive laboratory efforts, often yielding less than-optimal results.
This study focuses on the innovative design of polymeric materials for membrane-based gas separations. To streamline the material discovery process and guide synthetic procedures, we employ the online informatics platform Polymer Genome. This platform is utilized to predict the pure gas separation performance of novel polymers of intrinsic microporosity (PIMs) for CO2, N2, and CH4. In addition, we present the synthesis of a novel bio-derived ladder polymer inspired by PIMs, using ellagic acid as a biomolecule feedstock. Our findings indicate that while Polymer Genome predictions are not perfectly accurate, they offer sufficient reliability. Furthermore, we investigate the CO2/CH4 pressure-dependent permeabilities and solubility uptakes for the materials of interest. Finally, we employ the dual-mode sorption model, extending our study to CO2 and CH4 diffusivity at varying pressures.M.S.Chemical and Biomolecular Engineerin
High-efficiency multi-drug functional profiling in a microfluidic device towards personalized predictions in pediatric leukemia
No full textAcute lymphoblastic leukemia is the most common form of pediatric cancer, and while treatment outcomes have improved dramatically over the past several decades for these patients, this same progress has not been achieved for those who experience relapse or refractory disease, particularly in T-ALL. To further improve outcomes for these patients, new strategies need to be devised to both identify patients most at risk of relapse a priori to treatment exposure and to optimally match them with emerging new therapeutic options. To that end, in this thesis, we describe the development of an efficient microfluidics-based assay for functional evaluation of candidate drug combinations in pediatric ALL. In Aim 1, we first describe the development and characterization of the system and demonstrate its capabilities using leukemia cell lines. Next, in Aim 2, we evaluate response to standard induction therapy across a set of diagnostic ALL patient samples, and investigate the predictive value of the assay through correlation to clinical outcome metrics. Lastly, as a final proof-of-concept, we use the developed and validated system to evaluate an experimental treatment regimen in a set of patients who did not respond well to standard regimens. In sum, the results of this work have provided a new methodology for higher-order combination drug screening in limited sample settings, investigated the predictive value of combination response profiling in the context of available clinical data, and finally, evaluated an experimental drug combination across a set of samples from clinical non-responders.Ph.D.Biomedical Engineerin
Optimization of the Expression and Purification of Human Acid ß-Glucosidase in E. coli
No full textHuman Acid-ß Glucosidase (GCase) is a membrane-tethered lysosomal enzyme that is associated with two debilitating diseases, namely, Gaucher disease and Parkinson’s disease. To create a platform for studying the prevalent mutation of the enzyme associated with Parkinson’s disease, GCaseE326K, which is difficult to prepare using conventional methods, an engineered GCase enzyme was expressed and purified in E. coli cells. Several variables were tested to optimize the expression and purification of this designed protein, such as the E. coli bacterial cell lines, cell incubation temperatures, growing broths, and purification methods. The activity of the enzyme was tested using an enzymatic activity assay. With more optimization it is feasible to produce the GCase enzyme using this method. The biochemical and structural characterization of this enzyme and the effect of its activating substrate Saposin C (SapC) are vital to gaining a broader understanding of its associated diseases and potential treatments and therapeutics.UndergraduateChemistr
Bridging the Gap: Toward a National Framework for Local Climate Adaptation
No full textMCRP applied research paperAs adverse impacts of global climate change intensifies, local governments in the United States are increasingly responsible for climate adaptation planning. However, they operate without a cohesive national framework on climate change adaptation strategies. This paper evaluates climate adaptation plans of selected U.S. cities against fundamental principles derived from academic and policy literature. The study reveals wide disparities in the quality and comprehensiveness of city-level strategies for climate adaptation. The analysis underscores the urgent need for a standardized yet flexible national adaptation framework. Five policy recommendations are proposed to support more effective, equitable, and resilient local planning: establishing a national climate action planning framework, embedding financing mechanisms, planning for uncertainty, integrating equity and justice, and enhancing coordination across agencies. Together, these steps can help bridge the gap between fragmented local efforts and the scale of climate risks facing U.S. communities
Maximizing SSD Bandwidth Efficiency Through In-Storage GPU Integration
No full textThis study introduces a novel hybrid SSD-GPU architecture aimed at meeting the growing need for accelerated data processing. Our approach is designed to enhance overall efficiency by minimizing data transfer overhead and boosting storage bandwidth. We specifically focus on deploying a modified SBIOS algorithm, which harnesses internal parallelism to mitigate cross-block latency associated with write requests.UndergraduateComputer Scienc
Spatial AI analysis of single-cell image-based omics data
No full textA diverse array of spatial learning methods tailored for analyzing spatially resolved singlecell multiplex omics data are introduced. Initially, SpatialViz pipelines are introduced, which serve as foundational tools for analyzing disease and health dynamics at the single-cell level within their spatial contexts. By expanding the repertoire of protein targets, SpatialVizScore and SpatialVizPheno enable the uncovering of multiple interaction types across patient tissues. Subsequently, the integration of metabolomic and protein data is presented as crucial for understanding metabolomic changes at the single-cell level, offering insights into cellular metabolism and its implications for disease pathogenesis and treatment response. As the adoption of spatial omics technologies increases across diseases, integrating multi-omics data becomes imperative for deciphering underlying biological processes. Lastly, graph-based deep learning pipelines are described for analyzing single-cell and subcellular omics data, offering a powerful framework for capturing complex relationships within spatially resolved datasets. These pipelines demonstrate the potential to extract disease biomarkers and predict treatment responses accurately and help with cases of missing or lower-quality multiplex stains.Ph.D.Electrical and Computer Engineerin
Remake Remodel
No full textDiscussion portion of Lost in the Stacks, episode 638. Features a discussion about the subject liaison model of academic library service, and how it is changing at Georgia Tech.Discussion portion of Lost in the Stacks, episode 638. Features a discussion about the subject liaison model of academic library service, and how it is changing at Georgia Tech
Utilizing Combinatory Adjuvant-Loaded Chitosan-Derived Nanoparticles for a Joint SARS-CoV-2/Influenza Vaccine
No full textAs new pathogens arise and spread, potentially on the level of a global pandemic, the need for vaccines that target these pathogens continues to grow. Subunit vaccines are a standard method of protecting the body from these pathogens by administering a relevant protein from the said pathogen. These vaccines require the assistance of adjuvants to produce a sufficiently strong immune response, which can be fine-tuned to further increase efficacy by more accurately representing the pathogen of interest. To ensure that antigen and adjuvant do not simply disperse away from the vaccination site, nano- and microparticles are typically used as carriers. These PLPs are made of less reactive materials, leaving room for immunostimulatory functionalization.
Our primary hypothesis is that “chitosan-derived nanoparticles can serve as a flexible and functional adjuvant-carrying system as part of a vaccine.” In this work, we fabricated chitosan nanoparticles and presented them alongside combinatory nucleic acid adjuvants to create a proof-of-concept vaccine for SARS-CoV-2 and H5N1 influenza. First, we assessed various combinations of nucleic acid-derived adjuvants in vitro, monitoring the response in cell cultures containing either GM-CSF or FLT3L to arrive at a formulation we believe would be ideal for this purpose. We have also chosen to characterize cytokine-secreting cells using flow cytometry to assess potential sources of the immunological differences we observed between cultures. Lastly, we administered subunit vaccines in vivo utilizing our combinatory adjuvant-loaded chitosan NP systems to determine efficacy.Ph.D.Bioengineerin