In this work, we focus on multiscale modeling and control of a seeded batch crystallization process used to produce ibuprofen crystals. For the modeling of the crystal growth process, we consider kinetic Monte Carlo (kMC) simulations comprising of molecule adsorption, desorption, and migration type microscopic surface events. To account for growth rate variability, we propose a model for growth rate dispersion (GRD), based on available experimental data, which will be applied at the individual crystal growth level in the kMC simulations. Finally, a model predictive controller (MPC) is developed in order to control the crystal size distribution of ibuprofen in the batch crystallization process and the MPC closed-loop performance is compared against constant temperature control (CTC) and constant supersaturation control (CSC) policies. The proposed MPC is able to deal with the constraints of the control problem, in addition to minimizing the spread of the crystal size distribution in a superior fashion compared to the other control methodologies, which improves the crystal product quality at the end of the batch.

Hydrogen is proposed by many to be the fuel of the future as it is the key ingredient in a transition from a fossil fuel-based economy toward a zero carbon emission and sustainable energy economy. Hydrogen can serve as an efficient energy carrier for hydrogen-based technologies (e.g., fuel cells and hydrogen internal combustion engine) and lead to substantial reduction of greenhouse gas emissions and great environmental benefits. Hydrogen can be produced by a variety of technologies (e.g., steam methane reforming (SMR), coal gasification, biomass gasification, electrolysis, partial oxidation, solar thermal cracking) from fossil (e.g., natural gas), non-fossil (e.g., biogas) and non-carbon (e.g., water) sources, which highlights the great potential and flexibility of a hydrogen-based economy. Additionally, hydrogen is a key feedstock for the petroleum refining and fine chemical manufacturing industries. With current state-of-the-art technology, hydrogen is produced almost exclusively from fossil fuels by SMR. At SMR-based hydrogen plants, the reformers are the most expensive equipment in terms of the maintenance and operating costs, and thus, even a small improvement in the reformer thermal efficiency to lower operating costs of the reformer without compromising the expected service life of the reformer is expected to allow the plants to achieve a significant profit.

Motivated by these considerations, a systematic framework for creating and simulating a computational fluid dynamics (CFD) model for an industrial-scale reformer at an SMR-based hydrogen plant and, subsequently, a framework for designing and evaluating a real-time furnace balancing scheme are developed in this dissertation. Specifically, a CFD model for an industrial-scale reformer is created in ANSYS Fluent, which is used to improve our understanding of the physiochemical processes in the tube side and the furnace side of the reformer as well as their thermal interactions during the catalytic conversion of methane. Then, a furnace balancing scheme is developed to optimize the reformer input at the nominal total furnace-side feed (FSF) flow rate that minimizes the inherent variability in the outer tube wall temperature (OTWT) distributions along the reforming tube length. Subsequently, a statistical-based model identification is developed to create a computationally efficient and robust model for the OTWT distribution as a function of the FSF distribution, total FSF flow rate and interactions among neighboring reforming tubes so that the optimized reformer input can be identified in real-time. Finally, a real-time furnace-balancing scheme is developed to optimize the reformer input such that the reformer thermal efficiency is maximized without compromising the expected service life of the reformer.

Misogynist extremism has increasingly become a topic of concern for many researchers. Most research on misogynist extremism examines the online context that encourages the development of related ideologies. Using secondary data from a multitude of government and non-government sources, such as the Profiles of Individual Radicalization in the United States (PIRUS) and the American Community Survey (ACS), I conduct logistic regression and qualitative comparative analysis (QCA) to examine the regional factors that are associated with misogynist extremist presence in a county. I argue that gender-related, race-related, and economic regional factors help cultivate a hypermasculine social climate prime for misogynist radicalization. In the full regression model, only militia group presence is associated with misogynist extremist presence in a county. No sets were identified as true according to the inclusion ratio for sufficiency or the coverage of the outcome measure under QCA. Overall, there is some support that gender-related regional factors may contribute to misogynist radicalization.

Argumentation is known to improve understanding of primary literature and aid in achieving the goals of undergraduate biology education. However, argumentation has not traditionally been taught in the undergraduate classroom setting, leading to deficits in this skill among students. To support students to develop strong argumentation skills, and understand where students are struggling when constructing written arguments, it is essential to evaluate how students construct their arguments. Hence, my project aimed to develop a framework to answer the following questions: (1) What is the structure of students’ arguments: how much of their argument is addressing a claim, evidence, and mechanism? (2) How often do students include relevant and correct information in their arguments? (3) What are the common ways that students describe evidence in their written arguments? (4) Does argument quality improve after engaging in peer-review? Using a Toulmin framework, we assessed 240 arguments, written by students who had taken an undergraduate genetics course at the University of California, San Diego, over two years. Current findings reveal that students’ arguments were largely describing evidence, with a low inclusion of biological mechanisms to explain the data. Additionally, the majority of students described evidence as general trends instead of conducting data analysis. Assessing correctness and relevance of statements that make up an argument was not sufficient to reveal the struggles students have with writing scientific arguments. Also, our data suggested that a peer-review process might benefit students with more careful instruction.

Ion channels are integral membrane proteins facilitating permeation of ions through aqueous pores across cell membranes. They are essential to numerous physiological processes such as nerve impulses and heartbeats and are key therapeutic targets due to their involvement in various diseases. This dissertation employs advanced computational techniques to investigate the structure and function of ion channels, specifically focusing on the human (denoted by the prefix “h”) voltage-gated sodium channel hNaV1.7, voltage-gated potassium channel hERG, and small-conductance calcium-activated potassium channel hSK2, which are critical therapeutic targets for pain management and the treatment of cardiac arrhythmias. A brief overview of ion channels and lipid membranes as well as their experimental and computational studies is provided in Chapters 1 and 2, followed by three specific ion channel related stories in Chapters 3 through 5 and finishing with closing remarks and future outlook in Chapter 6.The first story focuses on the human voltage-gated sodium channel hNaV1.7, a prime target for pain therapy due to its predominant expression in peripheral nociceptive neurons. We explored the interaction between the hNaV1.7 channel’s voltage sensing domains (VSDs) II and IV in different conformational states and the tarantula venom peptide protoxin-II, a potent and selective inhibitor of the channel. Our computational models revealed that protoxin-II exhibits state-specific binding modes to different states of the channel’s VSDs, illuminating the molecular basis of its inhibitory effects through enhanced hNaV1.7 channel deactivation and inactivation. Additionally, we identified key channel’s residues involved in the toxin binding and their energetic contributions. These findings are crucial for the rational design of novel pain therapeutics with improved selectivity and potency. The second story shifts focus to the human voltage-gated potassium channel hKV11.1, also known as hERG (human Ether-à-go-go-Related Gene), a vital component of cardiac repolarization and a notorious drug anti-target. This makes the hERG channel a primary target for assessing the safety of new drugs due to association of its inhibition with drug-induced arrhythmias. Utilizing AlphaFold2, a state-of-the-art deep learning tool trained to predict protein structures from amino acid sequences, we modeled the elusive inactivated and closed conformations of the hERG channel not available experimentally. Molecular dynamics and drug docking simulations validated these models, aligning closely with existing experimental data. Drug docking simulations revealed state-specific drug interactions, underscoring the critical importance of the inactivated state for drug binding and the potential for drug entrapment in the closed state. Analysis of interaction networks across different conformational states of the hERG channel provided insights into potential molecular mechanisms underlying state transitions. These findings are pivotal not only for understanding the functioning of this crucial channel but also for predicting drug-induced arrhythmias from drug chemical structures and thus developing cardiac-safe medications. The third story explores the regulation of the human small-conductance calcium-activated potassium channel hSK2 by phosphatidylinositol 4,5-bisphosphate (PIP2) lipid in cardiac cells, aiming to enhance therapeutic strategies for cardiac arrhythmias like atrial fibrillation. Using optogenetic constructs to deplete PIP2, our collaborative study demonstrated that PIP2 depletion reduces hSK2 current in both cultured Chinese hamster ovary (CHO) cells and rabbit ventricular myocytes, underscoring PIP2's crucial role in hSK2 channel activation. Homology modeling and molecular dynamics simulations revealed the dynamic binding sites and mechanisms of PIP2 interaction with hSK2 channels, identifying key residues like R395 and E398 involved in channel activation by PIP2. This detailed analysis provides insights into the molecular mechanisms regulating cardiac excitability, offering potential targets for new anti-arrhythmic therapies. In summary, this doctoral dissertation delves into the intricate world of membrane proteins, in particular, ion channels, exploring their structural and functional intricacies including their interactions with the lipid bilayer environment, pharmacological agents, and toxin peptides. By employing cutting-edge computational approaches, we provided novel insights and strategies that hold promise for accelerating drug discovery and advancing precision medicine, ultimately aiming to develop more targeted and effective therapies for cardiac and neurological disorders.

Lipoprotein lipase (LPL) is a key enzyme responsible for the lipolytic processing of triglyceride rich lipoproteins (TRLs), with deficiencies in LPL known to cause familial chylomicronemia and mutations associated with hypertriglyceridemia. Upon synthesis, LPL is secreted into the interstitial spaces where it binds to GPIHBP1, an endothelial cell (EC) GPI-anchored protein. GPIHBP1 facilitates the transport of LPL across the capillary lumen where it functions in the intravascular processing of TRLs. This study leverages single cell RNA sequencing (scRNA-seq) to investigate the cell-type-specific expression of Lpl and Gpihbp1 across tissues. In our study we confirm the presence of known LPL synthesizing cells in mouse cardiac and adipose tissues, validating previous findings, and providing further support for their role in lipid metabolism. Additionally, our investigation highlights a novel observation: the expression of Lpl in the endothelial cells of skeletal muscle tissue. Most notably, these analyses uncover new insights into Lpl and Gpihbp1 expression in the choroid plexus of embryonic mice, unveiling a possible role of LPL-mediated intravascular lipolysis in the central nervous system. Overall, the study provides a comprehensive overview of Lpl expression across tissues known for robust LPL-mediated lipoprotein processing. These findings underscore the importance of understanding tissue- and cell-type-specific gene expression patterns to elucidate the complex regulatory mechanisms underlying lipid metabolism and its implications for human health.