UC Davis

Prostate cancer with distant metastasis (mPCa) is associated with poor prognosis. First-line treatment for localized PCa includes radical prostatectomy (RP) for high-risk disease. Many patients experience androgen receptor (AR)-associated increase of prostate specific antigen (PSA) levels in the serum, aka biochemical recurrence (BCR), increasing potential for mPCa progression. PCa cells that develop resistance to AR based therapies, along with chemotherapy, immunotherapy and radiotherapy, are associated with significant side effects and low overall survival. This dissertation assembles information linking the regulation of BCR to AR-associated metabolic pathways involved with metastatic progression. My goal was to identify metabolic pathways that disrupt BCR and determine potential targets of therapy that eliminate prostate cancer cells unaffected by AR inhibitors.In efforts to achieve this goal, I first conducted metabolomic analysis of prostate tissue from the tumors of 74 patients who underwent prostatectomy as treatment for localized PCa and correlated levels of metabolites with clinical and non-clinical factors. In this process, I identified that elevated levels of 2- hydroxyglutarate (2-HG) in the tumors reduced time-to-recurrence. Exogenous addition of D-2-HG, an enantiomer of 2-HG upregulated in PCa tumors from patients who experience BCR, increased the growth rate of LNCaP and C4 prostate tumor cell lines along with phosphorylation of Akt and ERK.I next studied targets of 2-HG that may affect the course of PCa progression, predominantly through use two castration resistant PCa (CRPC) sublines of hormone sensitive prostate cells (HSPCs)– (i) LNCaP-derived C4 cells, sensitive to AR inhibitors such as enzalutamide; (ii) CWR22-derived 22Rv1 cells, insensitive to enzalutamide. Collaborator-developed small molecule compounds were tested on cell lines of varying AR-sensitivity, including a benzimidazole-based molecule LLS139, observed to significantly decrease cell growth when treated intracellularly at its IC50. LLS139 treatment sensitized the more metastatically aggressive 22Rv1 cells to enzalutamide treatment compared to C4 cells while normal prostate-derived RWPE-1 and Normal Dermal Fibroblasts appear unaffected, highlighting potential in LLS139 as a treatment with reduced side-effects.RNA sequencing and qPCR validation highlighted several genes potentially contributing to mechanism(s) involved in the observed inter and intra differences between C4 and 22Rv1, such as upregulated GPNMB (glycoprotein non-metastatic B; aka HGFIN or osteoactivin)- a type 1 transmembrane glycoprotein expressed on the surface of cells or lysosomes. GPNMB was upregulated following LLS139 treatment of both C4 and 22Rv1 cells yet more significantly in 22Rv1 cells than with C4 cells, supporting drug validation.Proteomics assessment of co-immunoprecipitation experiments yielded Akt2 as a potential binding partner to LLS139. CCT12890 AKT2 inhibitor was employed with GPNMB siRNA transfections to help establish potential protein involvement. GPNMB knockdowns decreased GPNMB and reduced cell death, as observed through MTT experiments, significantly in 22RV1 cells but insignificantly in C4 cells. Treatment efficacy was significantly impacted by the Akt2 inhibitor for both C4 and 22RV1’s, appearing to mimic LLS139 treatment but observed to be less impactful on 22Rv1 than C4 cells. Combinations on inhibitor and siRNA proved fatal for both cell populations as no cell growth was observed. The above results indicate that LLS139 is likely a significant therapeutic agent in AR-insensitive PCa cells that target Akt2, operating through upregulation of the putative tumor suppressor GPNMB.

This dissertation consists of three chapters analyzing the consequences of populism in Europe. The first two chapters use Difference-in-Differences designs to study how populist parties in government affect liberal democracy. In the first chapter, I show that populist entry into government reduces liberal democracy scores—an effect driven entirely by right-wing populist parties. The second chapter disaggregates these effects, finding that right-wing populists in government specifically erode civil liberties and freedom of expression. Together, these two chapters establish that government status and host-ideology are important moderators of populism’s democratic consequences. In the third chapter, my co-authors and I examine elite-level party conflict and cooperation in the European Union. We show that growing support for populist and Eurosceptic parties has generated a new transnational cleavage that challenges European integration. Our findings show that elite-level, cross-national party conflict and cooperation is related to parties’ differing orientations towards European integration and populism, as well as their governing status. This dissertation demonstrates that populism can impact elite-level party competition and subsequently be linked to the erosion of democratic institutions. Together, this research offers new insights into the challenges facing democracy in Europe, contributing to the growing literature on how the emergence of populism has reshaped European politics, the strategic incentives that shape elite behavior, and the consequences these choices have for citizens and democratic institutions.

Many of the most well-known open problems in geometric topology involve the behavior of exotic smooth structures in dimension 4. One class of open problems in this field are the Wall-type stabilization problems, which ask about the behavior of exotic 4-manifolds under specified topological operations. For a chosen link homology theory, the associated skein lasagna module [MWW22] is an invariant of smooth 4-manifolds with robust gluing properties [MWW23]. We employ the skein lasagna module construction for Khovanov homology and study the effects of various forms of stabilization on these smooth 4-manifold invariants. Through the computations of these invariants for 4-manifolds relevant to external stabilization, we establish an isomorphism between Rozansky-Willis homology groups and skein lasagna modules of #k (S 2 × D2 ) with various null-homologous links in the boundary. Furthermore, using skein-theoretic 4-manifold invariants defined in [MWW24], we show that skein lasagna modules are capable of distinguishing pairs of exotically knotted surfaces even after an internal stabilization.

Fear, mistrust, and unclear responsibility continue to impede the adoption of autonomous and collaborative technologies. Embodiment, the way people perceive their body and actions, provides a framework for understanding human–machine interaction, with potential to reduce fear, foster trust, and clarify responsibility between users and systems. However, translating embodiment theory into practical evaluation tools for human–machine interaction has proven challenging, as existing metrics often yield conflicting results. This thesis uses psychophysics to quantify embodiment metrics, investigating their relationships during whole limb movements and in prosthetic tasks. Chapter 2 systematically varies visual, auditory, and haptic feedback to examine how four common embodiment metrics covary, revealing inconsistencies with earlier findings and suggesting that these measures may capture distinct constructs. Chapter 3 focuses on the sense of agency, the experience of authoring one’s actions and their outcomes, using a tightly controlled, forced-choice psychophysical paradigm. We test the validity of intentional binding and introduce a novel precision-based index, the temporal discrimination limen. Results show that intentional binding does not consistently reflect agency, whereas the temporal discrimination limen is selectively sensitive to volitional control. We conclude that the temporal discrimination limen is a promising implicit metric for agency and outline future work to validate its relationship with explicit measures and to examine the role of the supplementary motor area in volition and time perception.

Missing data is a ubiquitous problem in predictive modeling, but the majority of current methods are designed for statistical estimation rather than prediction, leaving the comparative behavior of imputation-based and non-imputational strategies in prediction-oriented settings poorly understood. This paper introduces 𝚖𝚟𝙿𝚛𝚎𝚍, an open-source R program for prediction under missing data in supervised regression. The package includes three complementary paradigms: imputation-based prefilling (𝚕𝚖_𝚙𝚛𝚎𝚏𝚒𝚕𝚕), pairwise available-case estimation (𝚕𝚖_𝚊𝚌) and the Tower technique (𝚕𝚖_𝚝𝚘𝚠𝚎𝚛), which in turn is based on the Tower Property of conditional expectation and aims to predict accurately. 𝚕𝚖_𝚙𝚛𝚎𝚏𝚒𝚕𝚕 supports four techniques, including complete-case analysis, MICE, Amelia and the missForest algorithm, ranging from list-wise elimination to model-based and nonparametric imputation approaches. We perform a rigorous empirical evaluation on five benchmark datasets with different sample sizes, missingness rates and properties of the response variable, using 5-fold cross-validation and four conventional predictive metrics. Our evaluation shows that the performance of a method depends heavily on the dataset characteristics and no single approach universally dominates all dataset combinations, thus motivating the necessity for a uniform benchmarking methodology. Practical guidelines for selecting a method depending on dataset features are also presented.The package is publicly available at https://github.com/matloff/mvPred.

Laser powder bed fusion (LPBF) additive manufacturing is a mesoscale process that uses small scale processing to create large scale parts. During this process, unconventional microstructures can form, which can have both favorable and adverse effects on part performance. Due to the complexity of the LPBF process and the large number of process parameters, controlling the resulting microstructures on a part specific basis has largely been impractical. While conventional simulation capabilities have shown promise for making predictions about the resulting microstructures, avoiding the time and monetary costs of printing and experimentally characterizing LPBF microstructures, these simulations are too slow for practical process tuning. This research aimed to develop a surrogate model, capable of making predictions about the microstructures produced by LPBF, fast enough to be used for process optimization.To achieve this, a framework for the creation of machine learning based surrogate models capable of making localized predictions about the statistical distribution of grain morphology metrics of interest was developed. This framework was then used to create models that can predict spatially defined grain size distributions for LPBF printed stainless steel, using a handful of thermal characteristics, calculated by a fast thermal model, that are representative of the LPBF process. The model is shown to not only be accurate in process regimes that are well represented in the training data, but also to improve on the accuracy of the training data by removing the stochastic error present in the data. The speed of the model, combined with the improved accuracy, allows for the prediction of the outcome of ensembles several thousand conventional microstructure simulations in a small fraction of the time. To ensure the framework and model are behaving as desired, a variety of methodologies for the analysis of the model were developed. These analyses provide insight into the behavior of the model with respect to the training data, the influence of the thermal characteristics chosen to represent the LPBF process, and the underlying assumptions used to create the model framework. The findings of these analyses support the validity of the framework and the usefulness of surrogate models created using it, showing promise for their future application to LPBF process optimization.

Soft matter assemblies, such as lipid bilayer membranes, liquid crystalline multilamellar phases, and cytoskeletal protein networks, are stabilized by collective interactions near the kBT scale, rendering them sensitive to environmental perturbations. This dissertation provides experimental evidence that structural reorganization of such assemblies is not governed by the strength of the environmental perturbation but by the physical mechanism of the stressor. Across two broad categories of environmental stressors, chemical and physical, three geometrically distinct model systems are investigated to assess this claim and capture evidence.In the first system, this dissertation reviews the theoretical framework connecting molecular packing geometry to membrane morphology before developing the pH-responsive bolaamphiphile GC18:1 as an experimental case study. When GC18:1 is inserted into DOPC giant unilamellar vesicles, it converts a chemically inert host membrane into a pH-responsive material. These synthetic liposomes bud inward at mildly acidic conditions and outward at basic ones, a direct consequence of the molecule's geometry shifting with its chemical state. This establishes a theoretical and experimental foundation for this dissertation in which the character of environmental stressors, here the molecular geometry, drives membrane remodeling.In the second system, concentric cylindrical multilamellar assemblies treated as lyotropic smectic-A liquid crystals, known as myelin figures, are subjected to osmotic stresses from solutes spanning molecular weights of 92 g mol-¹ to 10000 g mol-¹, with the osmotic pressure matched across solutes by adjusting concentrations. Small-molecule osmolytes produce no sur-face instability at any pressure tested, while larger macromolecular osmolytes drive periodic axial corrugations consistent with the Helfrich-Hurault elastic instability within seconds of exposure. This result demonstrates that the determining variable is the entropic, excluded-volume character of the macromolecular stressor: depletion forces generated at inter-myelin interfaces by solute-excluded coronas couple to the smectic's elastic instability mode, driving symmetry-breaking buckling that colligatively equivalent small-molecule solutes cannot produce.Ongoing work in the third system examines whether this excluded volume mechanism generalizes beyond lipid membranes to cytoskeletal protein assemblies. TActin, one of the most abundant proteins in the eukaryotic cell, is examined within dextran-rich aqueous two-phase system droplets stabilized by small unilamellar vesicles as Pickering agents, approximating cytoplasmic macromolecular volume fractions of 20-40%. Preliminary results reveal that actin preferentially partitions into these crowded droplets and that the dextran-rich environment promotes its assembly into a filamentous state, even in the absence of conventional polymerizing buffer conditions.These results thus far suggests a novel approach for synthetic cell engineering and biophysical design.Taken all together, the completed and ongoing work supports a unifying framework in which stressor character is the primary determinant of soft matter structural response, independent of stressor magnitude. Across multiple geometries and soft matter assemblies, these findings advance a physical basis for understanding and engineering environmentally responsive soft matter systems.

This dissertation studies interacting electron models for quantum matter from two perspectives with different approaches respectively. The first part of this text focuses on rigorously proving Lieb-Robinson bounds for continuous space integer quantum Hall systems perturbed by short range two-body interactions. A Lieb-Robinson bound is an operator norm inequality that implies a finite rate of information propagation in a quantum system, a low energy analogue to the relativistic limit given by the speed of light. Many results of this nature have been obtained for lattice systems, but over the last several years, progress has been made in identifying classes of continuous space models that admit propagation estimates. In this thesis, we prove Lieb-Robinson bounds for two-dimensional continuum fermions in a constant, perpendicular magnetic field using the strategy of Gebert et al. \cite{gebert:2020}, and use this result to prove strong continuity of the Heisenberg dynamics in the limit where inter-particle interactions occur in infinite volume. We then shift to a different framework, namely that of lattice-localized frames, to represent a class of interacting quantum Hall systems in continuous space, as introduced by Bachmann and De Nittis, \cite{bachmann2025lieb}. With this framework, we define a family of parametrized Hamiltonians and construct a continuum analogue of the Hastings-Wen quasi-adiabatic evolution, i.e. dynamics in parameter space that satisfy Lieb-Robinson bounds themselves. Moreover, we then directly apply this to prove quantization of Hall conductance for these models. The new results of these sections are based on works in preparation, \cite{hingorani2026lieb,hingorani2026quantization}. The second portion of this dissertation focuses on two specific materials and phases: magic angle twisted bilayer graphene (MATBG) and superconducting lanthanum nickel gallium-2 (LaNiGa$_2$). MATBG has garnered a tremendous amount of attention since works such as \cite{MB} and \cite{tarnopolsky2019origin}, which first suggested the existence of flat bands in the electronic band structure of twisted bilayer graphene: bilayer graphene with a relative twist between the layers. Astonishingly, this phenomenon occurs for specific twist angles dubbed magic angles. This quenching of kinetic energy sets the stage for electron-electron interactions to dominate the behavior of the material. The onset of Mott insulating states at several carrier densities as seen through peaks in inverse compressibility \cite{expts-b} drew our interest. In collaboration with Oitmaa and Singh, \cite{hingorani2022onset}, we study the SU(4) Fermi-Hubbard model as a candidate effective model for describing the low energy behavior of MATBG, where we uncover the onset of charge incompressibility and Mott gaps, indicating the existence of the insulating phase.Additionally, we investigate the superconducting pairing mechanism of LaNiGa$_2$, which has been highlighted as a potential time-reversal symmetry breaking (TRSB) material below its critical temperature due to experimental evidence for internal magnetization, \cite{Ghosh2020}, \cite{Hillier2012}, \cite{Sundar2024}. In our recent work, \cite{tsnb-1mxc}, we report the existence of a Hebel-Slichter (HS) peak in the nuclear-spin relaxation rate, an effect attributed to the divergence in the electronic density of states as the superconducting gap opens. Furthermore, we study models with triplet and singlet spin pairing respectively to ascertain a mechanism that reproduces a HS peak. In doing so, we find that two distinct superconducting gaps and TRSB may not coexist, which raises questions about whether LaNiGa$_2$ really does break time reversal symmetry or whether a different model is required to fully characterize this material in its superconducting phase.

Chronic kidney disease (CKD) is a debilitating condition characterized by a gradual decline in kidney function, resulting in decreased glomerular filtration, increased permeability of the filtration barrier, and deficits in secondary renal functions such as vitamin D activation and erythropoietin production. This functional decline leads to secondary complications, including the accumulation of uremic metabolites, chronic acidosis, systemic inflammation, vitamin D deficiency, and low hematocrit. Through these mechanisms, CKD exerts profound systemic effects on extrarenal tissues, particularly within the musculoskeletal and cardiovascular systems. This dissertation explores the specific effects of CKD on tendon, bone, skeletal muscle, and cardiac muscle, and evaluates the efficacy of uphill treadmill exercise as a therapeutic intervention using an adenine-diet-induced CKD rat model. We report the first experimental evidence of tendon dysfunction in CKD, providing a novel platform for mechanistic research. Furthermore, we found that while chronic treadmill exercise successfully maintained skeletal muscle mass, it exhibited limited effects on bone structure and function. Notably, exercise led to weaker tendons in CKD animals, further suggesting that renal disease fundamentally impairs tendon remodeling responses. Finally, CKD completely blunted exercise-induced cardiac hypertrophy in treadmill-trained rats, a maladaptation that appears driven by altered protein turnover. Collectively, this work elucidates the tissue-specific impacts of exercise in the context of CKD, highlighting the underlying mechanisms driving these responses and identifying critical targets for future therapeutic interventions.

Within the west-central Sierra Nevada, archaeologists have long debated the effects of the Medieval Climate Anomaly (MCA) on the indigenous culture, owing largely to the scant archaeological evidence produced during this period. More broadly, climate events are frequently included within arguments surrounding the reorganization of culture or their inexplicable decline. Within this region, the argument leans towards the former, however, this has led to further questions regarding the nature of this cultural reorganization. The previous cultural pattern of greater residential-than-logistical mobility, and a greater reliance on pine nuts than acorns as a storable staple seems to end around the same time as the droughts start. This dissertation examines the relationships among diet, mobility, and drought in the west-central Sierra Nevada over the last ~4000 years. It asks two related questions: how settlement structure, residential mobility, diet, and climate articulated through time, and how indigenous communities adapted to short- and long-term climatic shifts, especially drought. While human behavioral ecology theory guides the main hypothesis that water, if treated as a resource patch, was a limiting resource during the MCA, I used stable isotope analysis of paleobotanical, faunal, and human remains within a diachronic framework, together with regional archaeological and environmental evidence to address these questions. The paleobotanical analysis established a local isotopic baseline and demonstrated meaningful differences between angiosperm and gymnosperm resources, allowing the long-term importance of pine nuts to human diets to be more directly evaluated. The faunal analysis established baseline expectations for animal protein and local sulfur isotope ranges and further explored paleoprecipitation estimation from collagen. These lines of evidence were integrated with human carbon, nitrogen, and sulfur isotope data to evaluate dietary change, residential mobility, and the possibility of non-local movement through time. The results indicate that substantial social reorganization occurred during the Recent Prehistoric I period (RP I; 1100-610 cal BP), coeval with the MCA droughts. Dietary change was not instantaneous, but the shift toward greater use of angiosperms, and reduced reliance on gymnosperm resources was fully manifest by RP I. Sulfur isotope analysis suggests reduced residential mobility during this interval and does not support a simple replacement model, though some non-local individuals were present. Paleoprecipitation estimates derived from plants, fauna, and humans further suggest that drought-related environmental stress formed an important part of this sequence. Taken together, the evidence supports an interpretation in which drought acted as a catalyst for the initial reorganization of settlement, mobility, and diet, while later territoriality likely helped maintain reduced residential mobility into the Recent Prehistoric II period (RP II; 610-100 cal BP).