Browsing by Author "Kim, Seonah, committee member"

Now showing 1 - 4 of 4

Embargo
Data-driven strategies for organic structure-property and structure-reactivity relationships
(Colorado State University. Libraries, 2024) Santhanalakkshmi Vejaykummar, Shree Sowndarya, author; Paton, Robert, advisor; Prasad, Ashok, committee member; Kim, Seonah, committee member; Nielsen, Aaron, committee member
The prediction of molecular properties plays a pivotal role in various domains, from drug discovery to materials science. With the advent of machine learning (ML) techniques, particularly in the field of cheminformatics, the prediction of properties for small organic molecules has witnessed significant advancements. This document delves into the diverse machine-learning strategies employed for the accurate prediction of properties crucial for understanding molecular behavior. In Chapter 1, I offer insights into the evolution of data-driven modeling through Quantitative Structure-Property Relationships (QSPR), highlighting promising advancements in utilizing chemical features to construct predictive models for molecular properties. In Chapter 2, I delve into the primary stage of modeling, focusing on data collection for predictive tasks. I illustrate how the integration of automation and computational tools' advancement can construct modular workflows for FAIR (Findable, Accessible, Interoperable, and Reusable) chemistry. This approach aims to enhance the usability and reproducibility of scientific data. In Chapter 3, I emphasize leveraging computational tools to access high-level data for small organic molecules. I showcase the creation of a novel metric for assessing organic radical stability, utilizing a comprehensive chemical database of radicals. This involves employing straightforward physical organic descriptors, namely fractional spin, and buried volume, computed through systematic computational workflows. In Chapter 4, I explore the progression of graph-based models designed to forecast molecular properties, specifically Bond Dissociation Energy. Additionally, I conduct a thorough examination of two particular applications pertinent to pharmaceutical and atmospheric chemistry. I demonstrate that utilizing a minimal number of molecules from the relevant chemical space can notably enhance large-scale machine-learning models. Finally, in Chapter 5, I combine the developed tools from Chapters 3 and 4, to perform goal-directed molecular optimization in identifying novel radicals for aqueous redox flow batteries using graph neural networks (radical stability, redox potentials, and bond dissociation energy) and reinforcement learning. This de novo molecular optimization strategy has successfully identified 32 new radical candidates. By amalgamating insights from diverse studies, this dissertation endeavors to offer a comprehensive grasp of how machine-learning strategies are transforming the terrain of molecular property prediction.
Open Access
Engineered co-crystals as scaffolds for structural biology
(Colorado State University. Libraries, 2022) Orun, Abigail R., author; Snow, Christopher D., advisor; Ackerson, Christopher, committee member; Kim, Seonah, committee member; Ho, P. Shing, committee member
Biomolecules, like protein and DNA, serve as the foundation of life. The structure of biomolecules can give insight to their functions. X-ray crystallography is a cornerstone of structural biology, revealing atomic-level details of macromolecular structures. Even with advances in X-ray diffraction technology, haphazard and tedious crystal preparation remains the bottleneck of routine structure determination. An alternative to the crystal growth challenge is a scaffold crystal. Hypothetically, if one had a high-quality crystal already prepared with large enough pores for diffusion of a macromolecule, a biomolecule of interest could join the scaffold crystal for scaffold- assisted X-ray diffraction. An ideal scaffold crystal must be highly porous for guest addition, modular for installation of various guest molecules, and robust in changing solution conditions. A crystal with guest anchoring sites for post-crystallization guest addition may provide a high-throughput technique for guest DNA-binding protein structure determination. The overarching goal of this work is to design a novel scaffold crystal capable of scaffold-assisted X-ray crystallography. The scaffold crystals we designed are co- crystals of DNA and DNA-binding protein. In the co-crystal, the DNA serves as the anchoring point for guest DNA-binding guest targets while the protein acts as connective tissue to hold the DNA structure together. The scaffold co-crystal we engineered, Co-Crystal 1 (CC1), is the first example of a porous host crystal for DNA-binding guests. Ultimately, the expanded co-crystals may serve as a revolutionary figurative "lens" for routine structure determination. In addition to scaffold crystal development, we advanced methods to enhance scaffold stability and solution-independence, thereby augmenting the bioconjugation toolkit for crystals containing stacking DNA-DNA junctions. Specifically, we optimized a known bioconjugation technique, carbodiimide chemical DNA ligation, templated by crystals with stacking DNA junctions. Furthermore, crystal crosslinking chemistries were optimized to provide crystal strength at both the nanoscale and the macroscale. Post- crosslinking, co-crystal nanostructures were preserved as assessed using X-ray diffraction and co-crystal macrostructures were bolstered in harsh solution conditions. The crosslinking chemistry and protocol guidelines may advance the progress of DNA crystals and protein-DNA co-crystals utility in biomedical applications and structural biology. We are on the cusp of using designed co-crystals to host guest DNA-binding proteins for structural biology, bio-sensing, and bio-therapeutic delivery. Successful engineering of a designed porous co-crystal will open numerous application possibilities and scientific questions. For example, a future study could focus on quantifying guest protein diffusion rates and adsorption strength inside the porous scaffold crystals. The technology presented here may advance the study of DNA-binding proteins and advance our understanding of key proteins for cancer and disease.
Open Access
Non-equilibrium states of disordered systems: from low-frequency properties of glasses to distribution function of active Ornstein-Uhlenbeck particles
(Colorado State University. Libraries, 2022) Shakerpoor, Alireza, author; Szamel, Grzegorz, advisor; Van Orden, Alan, committee member; Kim, Seonah, committee member; Gelfand, Martin, committee member
This dissertation focuses on stationary and dynamical properties of non-equilibrium systems of disordered matter. In particular, we discuss the correlation between the stability of ultra-stable to moderately stable amorphous solids and the structural fluctuations of the elastic field at low frequencies. We report a strong correlation between the stability and the structural homogeneity which we demonstrate numerically through the calculation of local elastic moduli of the solid. Notably, we do not identify any significant length scale associated with elastic correlations which bears specific implications for the wave attenuation in amorphous solids. In the second part of the dissertation, we shift our focus to the disordered systems of active matter. We derive a formal expression for the stationary probability density function of a tagged active particle in an interacting system of active Ornstein-Uhlenbeck particles. We further identify an effective temperature in the probability density function which allows for the subsequent numerical validation of our theoretical results beyond the linear response regime. We show that the effective temperature defined through the violation of the Einstein relation (or equivalently the fluctuation-dissipation theorem), can predict the tagged active particle's density distribution. Lastly, we derive theoretical expressions for the stationary probability density distribution and the current of a non-interacting active Ornstein-Uhlenbeck particle in a tilted periodic potential. We demonstrate the quantitative agreement of these expressions with our numerical results for small to moderate correlation times of the colored-noise. We further explore the dependence of the diffusive motion on the strength of tilting force. We observe a giant enhancement in the diffusion of the particle which becomes more pronounced with increasing the persistence time.
Open Access
The development and implementation of a hybrid rocket motor thrust stand to investigate the relationship between combustion chamber pressure and graphite rocket nozzle erosion in hybrid rocket motors
(Colorado State University. Libraries, 2022) Kronwall, Matthew, author; Windom, Bret, advisor; Marchese, Anthony, committee member; Kim, Seonah, committee member
Rocket motors frequently implement carbon-based nozzle inserts to insulate the motor from the heat produced by combustion. Over time these inserts will erode due to oxidation at the surface wherein oxidizing species found in the combustion products react with the carbon to form carbon monoxide. It has been shown that the largest contributors/oxidants to erosion are H2O, CO2, and OH, due to their high concentrations within the exhaust products and the low activation energy needed to react with the carbon surface. As such, a clear understanding of the rate of oxidation, or erosion, is critical to rocket motor design. Previous research has modeled many of these characteristics, yet this has largely been limited to solid rocket motors with combustion chamber pressures greater than 6.9 MPa. Earlier studies have asserted that combustion chamber pressure has a linear effect on erosion rates, but it is unclear whether this linear assumption can be extrapolated to lower chamber pressures. This research lays the foundational work to explore the relationship between combustion chamber pressure and erosion rates at pressures below 6.89 MPa. Based on the numerical modeling and rocket motor test firings described in this study, preliminary findings indicate that this linear assumption may not hold at combustion chamber pressures below 3.4 MPa. Initial numerical modeling shows a non-linear increase in boundary layer thicknesses as combustion chamber pressures fall below 3.4 MPa. It is postulated that thicker boundary layer slows the diffusion of the oxidizing species to the surface thereby decreasing the rate of erosion. Thus, the modeled results suggest a non-linear relationship between nozzle erosion and pressure may be present at lower chamber pressures. Moreover, pure hydrocarbon fuels generate high fractions of key oxidizing species (H2O, CO2, and OH) in the product stream and the impact of these fuels on carbon nozzle erosion has remained largely unexplored. A hybrid rocket motor test stand (HRMTS) was developed to perform test fires of a HTPB-N2O hybrid motor at chamber pressures between 2.07 MPa and 4.83 MPa. Supplementary research was carried out that explored hybrid motor injectors and their effects of combustion instabilities. Major milestones included, implementation of a new semi-autonomous LabVIEW VI, creation of a MATLAB model that predicts motor performance, design and manufacture of a modular hybrid rocket motor, and the development of a secondary model that uses gathered test data to predict transient throat diameters. Furthermore, the predicted nozzle erosion was validated with the measured nozzle surface geometry pre and post-test fire through the utilization of a coordinate measuring machine (CMM). Initial results show that, despite the non-linear boundary layer growth, a linear relationship between combustion chamber pressure and nozzle erosion may still be true for chamber pressures below 6.89 MPa. Testing also illuminated correlations between combustion stability with injector pressure and nitrous oxide phase, for which, poor oxidizer vaporization and injector pressure dramatically decrease combustion stability and motor performance.