UC Irvine

Alzheimer’s disease is a progressive form of dementia with no effective treatments. Episodic memory deficits are a defining cognitive symptom of Alzheimer’s disease, which may be related to changes in the medial temporal lobes (MTL). The MTL facilitates episodic memory function and is particularly vulnerable to Alzheimer’s disease pathology. Both sleep and physical activity are linked to episodic memory function, including pattern separation operations reliant on the MTL. Evidence in young adults suggests that sleep and exercise are each independently linked to mnemonic discrimination (behavioral correlate of pattern separation), but it is unclear whether they interact to do so, and through what mechanisms. Whether these relationships are altered in older adults at-risk for Alzheimer’s disease is also unknown. To fill this knowledge gap, a series of studies were conducted. In Chapter 2, self-report measurements of sleep, physical activity, and cognitive function were used to statistically model the mediating effect of sleep in the physical activity-cognition relationship. In Chapters 3 and 4, high-density electroencephalography was recorded during overnight sleep to measure the expression and coupling of non-rapid eye movement (NREM) sleep oscillations, a memory task assessing mnemonic discrimination was administered prior to and following sleep to measure sleep-dependent memory consolidation, and questionnaire data assessing physical activity frequency and duration was collected from 36 older adults. High-resolution resting-state functional magnetic resonance imaging data previously collected from this same participant cohort was analyzed to derive measures of resting-state network modularity and MTL centrality. Chapters 3 and 4 addressed three specific aims: (1) whether local expression and coupling of NREM sleep oscillations are associated with sleep-dependent memory consolidation (Chapter 3), (2) whether network modularity and measures of MTL centrality are associated with sleep-dependent memory consolidation (Chapter 3), and (3) whether physical activity is associated with these same mechanisms (Chapter 4). This research provides novel insights into the neurobiological mechanisms supporting the relationships among sleep, physical activity, and memory consolidation in older adults. Findings from this dissertation could guide prospective physical activity and sleep-based interventional studies to improve memory function and inform public health recommendations to reduce Alzheimer’s disease risk by underscoring the importance of sleep.

Experimental observations indicate that the human immunodeficiency virus (HIV) can replicate in both the follicular and extrafollicular parts of secondary lymphoid tissues, HIV's primary replication site. The former is an immune privileged zone with low cytotoxic lymphocyte activity, and thus HIV replication is primarily concentrated there. Mathematical models and stochastic Gillespie simulations show that this compartmentalization potentially explains several seemingly counterintuitive observations. First, the observation of post-therapy viral decline rate's independence of cytotoxic T-lymphocyte (CTL) presence in simian immunodeficiency virus (SIV)-infected macaques, assuming CTL-mediated lysis significantly contributes to viral suppression. Second, the slow emergence of CTL-escape mutants during chronic infection even if CTL-mediated lysis is responsible for viral suppression. Heterogeneity in CTL activity, and consequently the selection pressure and infected cell population sizes, between the follicular and extrafollicular compartments can explain these findings. The effect of secondary lymphoid tissue compartmentalization is also examined in the context of non-escape scenarios, where the mutant HIV strain has a different infectivity to the "wild" strain, but is recognised by the CTL response. These findings highlight the importance of measuring viral populations separately in the extrafollicular and follicular compartments; using viral load in peripheral blood as an observable hides the heterogeneity between compartments that might be responsible for the particular patterns seen in the dynamics and evolution of HIV in vivo.

In a general-purpose multicore parallel multithreaded environment, multiple threads work simultaneously to finish a task faster. Usually, threads need to communicate with each other due to the need to share data or synchronize access to shared data. Communications commonly occur when threads need to wait for the computational results of other threads before continuing their own computations; all necessary data dependencies need to be met before subsequent computations take place.

A barrier is a synchronization construct that enforces a collective pause and data sharing at a given execution point between all participating threads of a parallel computation. No thread proceeds beyond a barrier until all other threads have reached it. However, in a general-purpose system where many processes compete for the available computational cores, a scheduler decides which thread gets a core to execute its next line of code. The barrier can easily become a performance bottleneck due to its global blocking nature; one preempted thread blocks the progress of all other threads that are waiting at a barrier.

This dissertation introduces a novel technique that changes the global nature of the barrier into a distributed data-driven synchronization model with non-blocking thread progression guarantees. The idea is to exploit the algorithm-based memory access patterns to extract peer-to-peer interthread communication and remove the explicit use of a barrier synchronization construct. Our proposed technique is experimentally validated. The results are promising and show considerable robustness in performance as opposed to their barrier-based algorithm counterparts.

This thesis is composed of two distinct projects. Although both projects are devoted to studying situations in which the eigenfunctions of certain Schrodinger operators exhibit interesting or exceptional behavior, the nature of the results as well as the tools used in each of the parts are drastically different. The first portion is dedicated to a particular spin-system model in which a type of many body localization is proved. We use the Multiscale Analysis to show that localization in higher spin systems occurs within a certain interval near the bottom of the spectrum. The second project demonstrates that a carefully constructed defect in specific kinds of periodic media, most notably multilayer quantum-graph graphene, can create an eigenvalue embedded in the continuous spectrum with a corresponding eigenfunction that has exceptional properties.

Query performance is critical in database-supported applications where users need answers quickly to make timely decisions. For decades, databases have relied heavily on query rewriting to optimize SQL query performance. However, with the current prevalent use of business intelligence and interactive visualization systems, purely relying on the rewriting capabilities inside databases is insufficient to optimize queries generated by those modern applications. On the one hand, different applications have various performance requirements. Some applications prioritize responsiveness, requiring queries to be executed within strict time constraints, and others may prioritize accuracy. Traditional database-centric approaches fail to exploit such information and adapt to diverse application requirements. On the other hand, developers and domain experts possess valuable insights into the data and query patterns specific to their applications. However, the query optimization techniques customized using domain knowledge can be infeasible inside databases.In this thesis, we focus on providing middleware-based query-rewriting techniques to help databases seize opportunities to optimize queries. First, for many applications with stringent response time constraints, such as interactive visualization systems, we propose a machine learning-powered query rewriting framework (called Maliva) to rewrite the queries with various options and help the databases generate efficient plans. Maliva leverages those expensive and high-accuracy query cost estimators to guide their rewriting process. By considering a pre-defined time constraint, Maliva judiciously explores different rewriting options and balances the query planning time and the execution time to find an efficient rewritten query that meets the time constraint. Second, in many cases, developers want to use their domain knowledge about the applications and datasets to rewrite queries for better performance. We propose a human-centered query rewriting solution (called QueryBooster) to provide users with an express and easy-to-use rule language to define rewriting rules. In addition, QueryBooster allows users to express their rewriting intentions by providing example query pairs. QueryBooster then automatically generalizes them into rewriting rules and suggests high-quality ones to the users. Finally, to lower the bar of users adopting the proposed rewriting framework, we implemented QueryBooster as middleware-based multi-user system to provide query rewriting between applications and databases as a service. Treating both the applications and databases as black boxes, QueryBooster requires no code modifications to them. To use the service, users only need to replace the database connector between the application and the database with a customized version provided by QueryBooster. The customized connector automatically intercepts application queries and sends them to QueryBooster to rewrite them based on user-defined rewriting rules. QueryBooster’s service model brings SQL query rewriting to a new paradigm where (1) users can easily formulate, control, and monitor query rewriting; (2) they can share rewriting knowledge and benefit from the wisdom of the crowd; and (3) they enjoy the non-intrusiveness security and pay-as-you-go convenience.

The transparent outer layer of the eye, known as the cornea, has a vital function inbending and directing light onto the retina, which is essential for achieving clear vision. Current methods of correcting refractive errors that are becoming increasingly popular include LASIK and PRK surgeries; however, these invasive techniques come with downsides as well such as permanent thinning of the cornea. This thesis proposes the use of a non-invasive technique called electromechanical reshaping (EMR) that can be applied to the cornea to induce curvature change. EMR was performed on New Zealand white rabbit globes using a concave platinum lens as the working electrode in a 3-electrode setup. The primary outcomes to be measured include the aforementioned curvature change of the cornea and the change in distribution of collagen orientation: to observe such changes the advanced imaging modalities of optical coherence tomography (OCT) and second harmonic generation (SHG) have been employed. The curvature data can be analyzed through a developed MATLAB approach that reconstructs the 3D surface of the cornea from 2D cross-sectional OCT images and sphere-fits the surface to extract the radius of curvature. The collagen structural change can be analyzed in FIJI software using OrientationJ to quantify the orientation of collagen fibers. These methods serve as a sound quantitative way to analyze the outcomes of EMR. This thesis found that the mean curvature increased from 5193.73 µm ± 597.09 µm to 5429.16 µm ± 736.08 µm following EMR application. Regarding collagen structural change, most of the eyes did not show a statistically significant difference following EMR application. Due to experimental error because of the lack of standardization of pressure, however, the corneal curvature did not consistently increase and the corneas did not resemble the radius of curvature of the platinum lens used in EMR (7.486 mm). Future directions include standardization of the pressure applied on the cornea by the platinum lens, simultaneous OCT imaging during EMR application, seeing if corneas post-EMR retain curvature changes, and transitioning to applying EMR on eyes intact in New Zealand white rabbit heads.

Our knowledge of the atmosphere has increased immensely in the last few decades because of high-resolution "storm-resolving" climate models. With these models, we can simulate atmospheric processes including deep, moist convection with detail previously not possible giving us a more accurate representation of storms, precipitation, and atmospheric waves. However, limits continue to constrain our understanding of the dynamics of the atmosphere. We presently lack the ability to run these new storm-resolving models (SRMs) for the durations we need to understand the cloud-climate feedback. Meanwhile, running these SRMs for any amount of time produces very large volumes of data which are difficult to analyze properly. This work leverages disparate machine-learning approaches in an attempt to break through these deadlocks. First, we implement feed-forward neural networks to replace the computationally expensive explicit convection calculations within the "Super-parameterized Community Atmospheric Model" (SPCAM) allowing us to run the model at a fraction of the original computational cost but with the same accuracy even when realistic geographic boundary conditions are included. Second, we use deep generative models to analyze and organize SPCAM output. This allows us to identify unique types of convection as well as convective storm anomalies within the data. A third outcome involves expanding on this unsupervised learning work to compare different SRMs - including uniform resolution global cloud resolving models - and quantify which have similar representations of the dynamics of the atmosphere. We find that even among high-resolution SRMs there are substantial differences in the type, proportion, and intensity of convection in representations of atmospheric dynamics. Fourth, we leverage these deep, generative machine learning models to make a novel metric of climate change and use it to better understand the physical mechanisms driving changes in extreme precipitation. We capture anticipated signals of global warming with minimal human intervention while showing the importance of the convection regime type to controlling the changing spatial patterns of heavy rainfall.

Due to the mechanosensitive nature of stem cells, intense research has also focused on the role of the cell microenvironment in directing cell fate decisions and stemness acquisition. It has been demonstrated that modulation of biophysical cues such as lateral confinement or substrate stiffness can enhance induced pluripotent stem cell (iPSC) and cancer stem cell (CSC) generation. While biophysical studies commonly involve modulating cell-cell and/or cell-ECM adhesions, a systematic study on the effect of perturbing cell adhesion during reprogramming has not been performed. Therefore, we propose to study the effect of a collection of cell-cell and cell-ECM associated genes, known as the Adhesome, on stemness acquisition of reprogramming cells, in the context of iPSC and CSC reprogramming.

Here, we report on a custom RNAi screen targeting 103 adhesome genes which demonstrated that adhesome expression is largely refractory to somatic reprogramming and revealed two novel top hits not previously associated with stemness acquisition: SHROOM3 and CSRP1. scRNA-seq was performed on reprogramming fibroblasts and A549 cells with SHROOM3 and CSRP1 knockdowns, respectively. Adhesome expression knockdown in both reprogramming models decreased PCP and EMT signatures, increased stemness acquisition, and, as revealed by CellChat inferred communication patterns, utilized similar cell-environment signaling during stemness acquisition. Furthermore, trajectory inference of the adhesome gene knockdown in both reprogramming models revealed novel pathways to stem-like states and cell fate transitions. Together, our findings support the adhesome as a novel target during somatic and oncogenic dedifferentiation for the development of more effective stem cell therapies and cancer treatments.

Activism in authoritarian countries conjures images of crowded streets and clashes with police. However, movement towards effective change can be made in quieter spaces through targeted initiatives. Issue focused activism is occurring in urban spaces in Kazakhstan responding to centralized urban development schemes and local issues such as green space preservation, residential displacement, and urban infrastructure development. Successful civic activism in Kazakhstan’s cities sharply contrasts with the concurrent arrests and detainments of protestors. This study focuses on the mechanisms of change by asking: how do citizens in Kazakhstan impact urban development processes and outcomes under authoritarianism? Political science scholarship asserts that there is little civic participation in illiberal spaces. Participation at the urban level is deemed inconsequential “low politics” with no expected impact on national political culture or policy. This study examines citizen effectiveness using a multiple case study comparison of development projects in three cities in Kazakhstan: Almaty, Astana, and Shymkent. The results of this research show that citizens and urban activists are intervening in centralized development plans in ways that help achieve desirable individual and community-level outcomes. Methods and modes of intervention vary per city, based upon the issues citizens are faced with, and their cultural, historical and institutional legacies. Despite diversity in the case studies, collectively these studies show that citizens have the ability to transform state-society relations in the context of entrenched institutional interests and legacies, in an authoritarian system. Collectively, the analysis highlights regional variations in governance despite extreme political and fiscal centralization. Variations in governance and citizens’ ability to elicit impact urban development processes are context specific. Divergent experiences of governance in the case studies was unprecedented given the shared governance arrangements, and their role in national economic development policy as signified through their status as Special Economic Zones. Variations in governance within a centralized sovereign state problematize state and regime types, and show new possibilities for how citizens can advance their interests at the local scale under these conditions. While agglomeration has concentrated governance, it has not diminished variations in governance which are elevated to the regional scale. Regional apparatuses struggle to govern newly agglomerated regions which have more diverse populations and geographies than pre-existing smaller towns and cities. The diversity of persons increases the range of potential interventions within regions. Successful, and even stalled, interventions in urban space provide frameworks and strategies that can be shared and implemented across cities with the potential to shift state-society relations and increase the role of citizens. Citizen-state clashes over national level politics have resulted in deaths and detainments, while the urban provides an arena for safe activism and democratic experimentation absent in Kazakhstan’s national level politics. Insights into the modes by which citizens organize in response to complicated, quasi-governmental, hybrid, public-private development agencies provide guidance and a network for urban citizens experiencing similar phenomena in different geographies. Transnational networks of support, expertise and funding represent globalization from the bottom-up (Kostovicova and Glasius, 2011). This is consistent with a stream of scholarship that sees this type of activism and organizing as the natural response to articulations of globalizations in local space (Levinson, 2002; Tsing, 2011; Hornberger and McCarty, 2012; Keck and Sikkink, 2014). Findings from Kazakhstan expand the geography of resources and participation in this activism, and exhibit new modes of political participation to respond to these pervasive governance arrangements.

This selection of four short stories is broadly representative of my development as a writer during my time as an M.F.A. candidate at the University of California, Irvine, between the fall of 2021 and the summer of 2023.