22689 research outputs found
Sort by
MULTI-SCALE IMAGING CHARACTERIZATION OF CYTOSKELETAL BIOMECHANICS
No full textCells sense mechanical signals from and mechanically interact with the microenvironment through cytoskeletons. In recent decades, advances in imaging afford us to study the dynamics and structures of cytoskeleton associated with health and disease. I dedicated my training to develop methods and use them to further characterize cytoskeletons. The findings uncovered by these methods might be incorporated in new treatments for diseases associated with aberrant cytoskeleton functions.
My investigations span various spatial scales. At the nanoscale, I examined the performance of two different approaches to super-resolution image reconstruction, the sparsity-based method and deep learning-based method. Using simulated datasets with adjustable emitter density and connectivity, I found the two approaches have their advantages in different performance metrics. The conclusion can guide researchers to computationally recover ultrafine structures of the cytoskeletons of normal and diseased cells without conducting cost-prohibiting super-resolution microscopy.
Next, at the sub-cellular (µm) scale, I studied the effect of nuclear mechanics in chemoresistant ovarian cancer. I devised image analysis workflow to examine two major regulators of nuclear stiffness, nuclear heterochromatin content and the nuclear envelope. I found that the integrity of nuclear envelope is compromised by combined paclitaxel and spleen tyrosine kinase inhibitor treatment, while heterochromatin levels did not have a significant effect in the killing of ovarian cancer cells. These results highlight the potential value of the nuclear envelope as a therapeutic target in combating chemoresistant cancer.
Finally, at the cellular (tens of µm) scale, I identified a unique mechanophenotype in human volar fibroblasts. Using live cell microscopy, I tracked a large population of volar and non-volar fibroblasts from patients under different pressure conditions. I found that volar fibroblasts are more motile and can undergo a discontinuous, high-velocity “slingshot” migration. Furthermore, the phenotypic difference between volar and non-volar fibroblasts persists under pressure. These findings suggest volar fibroblasts can generate higher force and remodel ECM more efficiently, providing insight into fibroblast-mediated skin homeostasis
From Rain to Vein: Unraveling the Subseasonal Processes Linking Climate Variability to Dengue in Sri Lanka
No full textThe mosquito-borne disease dengue is one of the most rapidly growing infectious diseases in the world. A major contributor to global disease burden, it primarily impacts the many who live in the tropical and subtropical regions of the world that are climatically favorable for its mosquito vectors (primarily the species Ae. aegypti). In fact, dengue incidence itself is closely associated with climate, which has spurred numerous global efforts to develop climate-informed early warning systems to predict dengue risk. Thus far, use of climate data in such systems has generally relied on assessing broad statistical associations between variability in climate (e.g., air temperature) and variability in disease incidence. However, what remains hardly explored is the potential added predictive signal within the intermediate processes that connect climate and dengue. In this dissertation we modeled one such chain of processes—the impact of climate on dengue via the climate sensitivities of Ae. aegypti breeding habitats—to assess the potential of such process-based modeling to inform dengue risk prediction efforts in Sri Lanka. We did so by first analyzing climatic variability in Sri Lanka at the mosquito-relevant subseasonal-to-seasonal timescale, finding strong meteorological associations with the Madden–Julian oscillation (MJO) that suggest potential for local MJO-based precipitation forecasting. We then assembled a mechanistic modeling pipeline that simulates meteorological impacts on vector container habitat water dynamics, which in turn informs a simulation of population dynamics of Ae. aegypti eggs, larvae, and pupae. Driving this model for three Sri Lanka cities with reanalysis/satellite-based meteorological data (MERRA-2/IMERG) yields simulated vector abundances that have some notable location-specific associations with observed dengue incidence. Driving this model with meteorological data from a forecast model (GEOS-S2S) yields comparable results in one city but not the others, attributable to systemic (potentially minimizable) biases in the forecast data. In sum, the assembled modeling pipeline shows initial promise for producing dengue-relevant outputs. While there is more work to be done on rigorously testing the modeling pipeline within a more expansive dengue risk assessment framework, our findings highlight the value in exploring the nuanced intermediate processes that connect climate and dengue
TEACHER INSTRUCTIONAL AND MANAGERIAL DECISION-MAKING: THE RELATIONSHIP BETWEEN DECISION-MAKING AND SATISFACTION
No full textSelf-reported teacher satisfaction has been decreasing over the past two decades, and prior research indicates that working conditions, including the extent of teacher influence over school decisions, are a significant predictor of teacher satisfaction. However, few studies have explored how influence over different types of school-level decisions may impact teacher satisfaction. This dissertation uses data from the 2015-16 National Teacher and Principal Survey and utilizes descriptive statistics, factor analysis, and multi-level modeling to develop a more nuanced understanding of the relationship between teacher decision-making and teacher satisfaction. First, descriptive statistics reveal that unsatisfied teachers are clustered in a relatively small percentage of schools. Second, a factor analysis of teacher responses on school decision-making indicates a two-factor solution that separates decision-making into instructional decisions (i.e., standards and curriculum) and managerial decisions (i.e., professional development, teacher evaluation, hiring, discipline, and budget) is preferable to a single-factor or three-factor solution. Third, multi-level modeling indicates that school-level instructional decision-making is a significant, positive predictor of teacher satisfaction. In contrast, after controlling for other teacher-level and school-level variables, school-level managerial decision-making is a significant, negative predictor of teacher satisfaction. Finally, introducing interaction terms between school-level decision-making and teacher-level administrative perception provides evidence of a significant interaction between instructional decision-making and administrative perception when predicting teacher satisfaction and holding all other variables constant. Specifically, the interaction indicates that increased school-level instructional decision-making is most strongly and positively predictive of increased teacher satisfaction for teachers with negative perceptions of their administrators. This research contributes nuance to the existing research on the relationship between teacher satisfaction and teacher decision-making and encourages more research into teacher instructional decision-making as part of an approach for improving teacher satisfaction in schools with large percentages of unsatisfied teachers
MULTI-OMIC APPROACHES TO UNDERSTAND TUMOR EVOLUTION DURING IMMUNE CHECKPOINT INHIBITION
No full textImmune checkpoint inhibitors (ICIs), have transformed cancer therapeutics across cancers, however a sizable fraction of patients eventually develop therapy resistance. Therefore, it is essential to identify which patients would attain the most benefit from ICIs. This has triggered many studies focused on understanding, at a molecular level, response and resistance to ICIs, leading to the identification of tumor-intrinsic and -extrinsic features that may predict therapeutic response. Additionally, liquid biopsies have gained increased interest in being utilized to assess response in real-time and guide clinical decision-making.
This body of work aims to understand ICI response by employing a multi-omic approach to identify features that may be indicative of response and utilize liquid biopsies as a means to dynamically assess tumor burden and ICI response during therapy. ICIs have variable effectiveness across cancer types and are often more effective in highly mutated tumors with an inflamed tumor microenvironment. This has led to interest in utilizing other therapies, such as epigenetic modulation, as a way to sensitize immunologically cold tumors to ICI. In our work, we employ genomic and transcriptomic analyses in two separate clinical trial cohorts of patients that received both epigenetic modulation and ICI. Through transcriptomic analyses, we show evidence that epigenetic therapy led to reprogramming of the tumor microenvironment, which may in turn sensitize some tumors to ICI. Through comprehensive genomic analyses, we were also able to identify several genomic features that were uniquely enriched in tumors that responded to combination therapy.
Additionally, we employed minimally invasive circulating tumor DNA (ctDNA) analyses in a separate cohort of patients with resectable gastroesophageal cancer receiving neoadjuvant ICI to dynamically assess if ctDNA kinetics were reflective of clinical outcomes. Through these analyses we showed that persistence of ctDNA during the neoadjuvant treatment course is indicative of inferior clinical outcomes. Importantly, ctDNA status was more informative in predicting cancer recurrence compared to pathological response assessments.
Collectively, this body of work highlights the potential of a multi-omic approach to identify patients most likely to attain response to ICI and provides evidence that longitudinal ctDNA assessment can capture therapy response at a molecular level. These findings have broad and immediate implications in the clinical care of individuals with cancer across tumor types and stages
THREE PROBLEMS IN GEOMETRIC ANALYSIS
No full textThis thesis has three parts. Although, the questions and techniques in each part
are independent of each other, broadly they all emanate from the theory of minimal
surfaces.
In the first part we address the construction of minimal surfaces as limit sets of
interfaces in the boundary reaction equation. Specifically, we establish an epsilon
regularity result for boundary reaction equation and show that the zero set of critical
points with bounded energy converges to a stationary rectifiable varifold. This is
based on [44].
In the second part we address the Steklov eigenvalue problem for compact manifolds.
We prove an estimate for the first Steklov eigenvalue for manifolds with positive Ricci
curvature and convex boundary, that improves upon the existing best known estimate
for this problem. This is based on joint work with Jonah Duncan [20].
In the third part we address Gromov’s band width inequality and Rosenberg’s
S1-stability conjecture for smooth four manifolds. Both results are known to be false in
dimension 4 due to counterexamples based on Seiberg-Witten invariants. Nevertheless,
we show that both of these results hold upon considering simply connected smooth
four manifolds up to homeomorphism. We also obtain a related result in the general
case. This is based on joint work with Balarka Sen [43]
A Healing Stage: Violence, Self-Knowledge, and Therapeutic Theater in Israel
No full textThis dissertation is a historical and ethnographic study of psychodrama in Israel, with a focus on its practice in the mixed Jewish-Arab city of Lod. The study is based on five years of research, including an intensive fieldwork period of sixteen months (October 2021–January 2023). Ethnographic fieldwork involved participant observation of both adult and children’s theater rehearsals over multiple school years. It also involved repeated semi-structured interviews with adult theater members and psychodrama clinicians. Archival research spanned sites across the United States and Israel, including the Archive of the University at Tel Aviv University, The Yad Vashem Library in Jerusalem, Ben Gurion University Medical Library, the Israeli Psychoanalytic Society, and the Harvard Countway Library.
The dissertation argues that, by positioning development as the ‘capacity to encounter otherness’ and the aim of therapy as ‘the capacity to approach old problems in new ways,’ psychodrama offers a pathway for imagining collective life in Israel/Palestine that departs from dominant ethno-national narratives. The dissertation demonstrates how psychodrama’s shift of the scene of healing from the clinic to the stage fundamentally unsettles core assumptions in trauma theory—from who is the subject of therapy to what constitutes repair. Rather than evaluate the efficacy of psychodrama as a therapeutic intervention, the dissertation takes up the figure of the child, both as a figure of thought and as a method, to develop an anthropological analysis of theater scenes. Each chapter of the dissertation traces how the fragmented texture of children’s knowledge generates new insights for self-knowledge and knowledge in a form of life: from the way the legal category of crimes against humanity cannot fully capture the way mass violence corrodes the concept of the human, to how elementary school children’s play re-conceptualizes what might count as politics. The dissertation argues that, in light of these findings, repair is neither a return to a previous state nor the capacity to narrativize one’s experience of injury. Rather, repair resides within the capacity to hold onto the possibility of seeing the new, particularly amid an existential fear that the possibility for newness is slipping away
THE ROAD TO SAFE SYSTEMS IN LATIN AMERICA: OVERCOMING BARRIERS, STRENGTHENING GOVERNANCE, AND DRIVING IMPACT
No full textBackground: Road traffic injuries (RTIs) remain the leading cause of death and disability worldwide for individuals aged 15 to 44. The Safe System Approach (SSA) positions road safety as a shared responsibility among all stakeholders. However, there is limited evidence regarding implementing this approach in policy and practice in Latin America.
Objectives: Through the lens of the SSA, this dissertation aims to (1) examine the design and implementation of civic-based sanctions for speeding, (2) develop a method to enhance the effectiveness of speed cameras by prioritizing vulnerable road users, and (3) identify the facilitators and barriers to implementing the SSA within a road safety technical assistance program.
Methods: This study uses (1) abductive analysis of documents and key informant interviews (KII), incorporating the framework of Ideas, Interests, and Institutions; (2) geospatial analysis to characterize speed camera coverage and identify RTI hotspots; and (3) thematic analysis of KII based on the components of the SSA.
Results: Multisectoral collaboration, essential to a Safe System approach, emerged as a key theme in all three studies. Another significant theme was the strategic use of crash and mortality data to shape priorities and establish legitimacy. Furthermore, data served not only as a tool for planning but also as an instrument to reinforce the legitimacy of road safety interventions. Political leadership (or its absence) greatly influenced implementing a Safe System approach. Across cities, road safety interventions started as small-scale initiatives (pilots) that were either scaled up or down based on political contexts, as described above. Overall, governance structures proved to be a central factor in determining the success of integrating or fragmenting the components of SSA implementation.
Conclusions: Finally, this study provides a pathway for understanding how road safety interventions emerge (chapter 1), evolve and become more effective (chapter 2), and become institutionalized in various governance contexts (chapter 3)
DATA CONFIDENTIALITY FOR ALL: NEW METHODS IN ATTACK AND DEFENSE
Full text linkFor the past half-century since the advent of modern cryptography, perhaps most distinctly demarcated by Diffie and Hellman in 1976, dramatic advances in cryptographic theory and practice have propelled us from a world where secrets were best hidden by obfuscation and hope to one where private, authenticated communication is the almost invisible backdrop of all electronic communication and where advanced cryptosystems secure previously unimaginable functionalities such as privacy-preserving computation and even novel digital financial systems. One connecting thread in this history is that cryptographic constructions and cryptanalysis have largely been done by, and for, cryptographers. This is perhaps most clearly evidenced in the folklore maxim that almost every computer scientist hears at some point: “don’t roll your own crypto.” Recent efforts to promote usability and transparency in cryptography are poignantly summarized in seminal critiques of the complexity and opaqueness of cryptographic tools. The lack of accessibility in both building and evaluating cryptosystems is far worse than just an inconvenience of consulting experts. As Rogaway famously described, cryptography is a mechanism of power – the power to control the flow of information, to protect data sensitive to us, and to interact with modern technology systems. Thus, accessibility is a matter of equity as much as it is one of efficiency. When viewed through this lens, the lack of accessibility to cryptography expertise can be clearly understood as a problem at the intersection of the societal and the technical. Throughout this work, we explore opportunities to leverage automation, familiar tools, and commodity platforms to develop novel cryptographic methods. In doing so, we seek to democratize access to cryptography knowledge, implementation, and evaluation. Technology alone cannot be expected to solve societal problems. However, only by including considerations of accessibility and usability into our development of novel methods can we hope to pursue efforts which serve to mitigate the impacts of these societal challenges
BACKGROUND (UN)PROMOTABLE: WHAT MAKES CHINESE HIGH SCHOOL STUDENTS STAY AND LEARN IN ONLINE BACKGROUND PROMOTION PROJECTS?
Full text linkIn the educational contexts of China's international college application landscape, students engage in extracurricular activities known as "background promotion projects" to enhance their admission prospects. These projects, offered by for-profit educational agencies, aim to develop students' cultural capital and provide competitive advantages. However, the effectiveness and implications of these projects have been largely unexplored. This dissertation attempts to advance our understanding of background promotion projects and develop strategies for improving student engagement and retention rates.
The first contribution of this dissertation is a comprehensive sequential needs assessment of students and instructors involved in these projects. Through a mixed-methods approach for students’ needs assessment, I investigated factors such as perceived value and instructional effectiveness, and analyzed their impact on student retention and engagement. The mixed-method study analysis reveals that perceived instructional effectiveness significantly influences students' intention to complete projects. For the qualitative needs assessment for instructors, I explored how instructors perceived their teaching experiences and challenges. The second contribution is the development of practical strategies to enhance instructional effectiveness and student retention in online background promotion projects. To address the challenges of low retention rates and inconsistent instructional quality, I explore the potential of AI-assisted tools, particularly focusing on MagicSchool.ai, to support instructors and improve student experiences. Finally, I discuss the pedagogical implications of AI integration in education, hypothesizing a shift towards more dynamic, personalized learning environments that foster critical thinking and authentic learning experiences
Contested Legacy: A Re-Examination of the Ideologies, Ethos, and Interventions of the Turkish Military
Full text linkThe military is one of the most consequential institutions in Turkey, shaping its founding, generating much of its political leadership, and repeatedly re-orienting its politics through a series of interventions. Given its historical significance, the military is deservingly one of the most studied institutions in the country. Despite extensive scholarship across diverse academic disciplines, studies of the institution present a remarkably consistent picture of the military as a unitary and monolithic actor, committed to a rigid set of Kemalist ideals. This thesis argues that this dominant scholarly view of the Turkish military is flawed, and by extension, has contributed to the failure of academics and policy makers alike to anticipate and comprehend key events in Turkey, including the failed 2016 coup. Using Turkish military interventions and major wars as markers of, and windows into, periods of political and ideological change, this thesis constructs a new periodization of Turkish political history, beginning with the late Ottoman Empire and culminating in the 2016 coup attempt. Re-examining much of the historical literature, including both primary and secondary sources, it argues that the military has historically possessed much more ideological diversity than generally assumed. Instead of a monolithic and unitary actor committed to strict Kemalist orthodoxy, it argues that the military has served as a theater of competition between robust and divergent political views inherent across the Turkish political spectrum. Moreover, it argues that these divergent ideologies played important roles in driving the actions and shaping the policies of the military, including within each coup