22689 research outputs found
Sort by
BRIDGING EXPERIMENTAL AND IN-SILICO METHODS FOR ADVANCED MAMMALIAN BIOMANUFACTURING OPTIMIZATION
No full textModern biotherapeutics production is dominated by mammalian cell lines due to their ability to produce complex biomolecules at scale while maintaining appropriate quality for treating human disease. Chinese Hamster Ovary (CHO) and Human Embryonic Kidney (HEK293) cells have emerged as the major players in producing monoclonal antibodies (mAbs) and recombinant adeno-associated virus (rAAV) gene therapy vectors, respectively. Despite this, biomanufacturing these therapeutics faces challenges such as inefficient cell medium optimization and a lack of understanding of biological effects during production. In gene therapy bioprocess engineering, the effects of cell media and process conditions are not well understood, leading to low titers and poor product quality. Optimizing cell culture medium for both mAbs and rAAVs is expensive and time-consuming, involving many wasteful experiments during development.
This thesis tackles some of the challenges in bioprocess development, focusing on mAbs and rAAV gene therapies. It explores bioanalytic methods and model-guided approaches for cell culture media design and for better understanding cell health during production.
First, we focus on rAAV vector production using HEK293 cells, highlighting medium composition and process optimization to achieve consistent vector genome (VG) titer and quality. We then explore cytotoxic challenges in rAAV production, particularly caspase-mediated apoptosis, and strategies to control cytotoxicity, demonstrating how inhibiting apoptosis can enhance product quality.
Next, we leverage thermodynamic models to study nutrient interactions in cell culture media, designing optimal feed media compositions while preventing precipitation. By collecting extensive data on amino acid solubility and activity coefficients, a functional group-based UNIFAC model was developed to ensure stable media formulations, applicable to mAbs, rAAVs, and other bioprocesses requiring high nutrient concentrations.
Then, we discuss an advanced modeling approach to simulate CHO cell culture by combining flux balance analysis (FBA), kinetic modeling and integrating 13C-labeled data to improve predictive accuracy. This hybrid model simulates CHO cell metabolism in batch and fed-batch systems. Finally, the the last chapter integrates machine learning with bioprocess development, proposing a data-driven approach using Bayesian Optimization (BO) for efficient cell culture media design. Combining machine learning with the thermodynamic models, and metabolic models from the previous chapters, a bioprocess-tailored BO framework was developed, contributing towards automated bioprocess optimization.
By addressing these interconnected challenges and leveraging bio-analytical and model-guided approaches, this thesis demonstrates how biotherapeutic production can be made more efficient. These modest contributions are made to advance bioprocess design with the ultimate goal of making these life-saving treatments more accessible
Hierarchy, class, race and PPE in an American hospital in the early days of COVID-19: What the pandemic stress test can teach us about building equitable health systems
Get PDFBecause hospitals are spaces where life and death are routinely at stake, social hierarchies, pressures, and cultural norms are heightened. This was particularly true in the early days of the COVID-19 pandemic. Examining the dynamics in that era can provide insight into the nature of race and hierarchy in hospital structures.
In the large literature on the experiences of hospital staff in the COVID-19 era, class and racial dynamics in hospitals are often sidestepped. In addition, the experiences of service staff such as environmental service workers and food service workers are severely under-represented.
Here, we explore hierarchy, class, race and risk in two hospitals in the city of Baltimore in the first months of the pandemic in 2020, through the lens of availability of PPE. We draw on a survey of 403 staff in two Baltimore hospitals, and semi-structured interviews with 57 of those staff. Respondents worked in a variety of roles, from administration to clinical to service staff.
A large majority of non-clinical service staff identified as Black, in contrast to a small minority of clinical staff with advanced degrees. The experience of access to PPE in the early pandemic differed across cadres of workers. Everyone in the hospital had to ration PPE, but many service staff felt that they were not prioritized in the same ways as clinical staff. PPE availability took on powerful symbolic resonance as a demonstration of how different cadres of workers were valued.
The COVID-19 pandemic threw social and class dynamics within the hospital into relief, shedding light on what so often ran below the surface. Thus, it could also potentially be an impetus to examine these fault lines, and to push hospital structures a bit more in the direction of justice
Understanding Experiences of Educator Burnout Through Reddit Analysis
Get PDFThis study investigates the issue of teacher burnout, a pervasive challenge within the educational sector, exacerbated by organizational and individual factors. Utilizing Bronfenbrenner's Ecological Systems Theory and the Dual-Level Social Exchange Model, the research explores the nuanced dynamics contributing to burnout, particularly through online platforms like Reddit, where educators share their experiences. The study employs qualitative methods to analyze Reddit posts, focusing on the lived experiences of teachers facing burnout and the support they receive through online communities. The findings reveal that teacher burnout is significantly influenced by inadequate organizational support, unrealistic expectations, and a lack of mentorship. The analysis highlights the role of social media in providing an informal support network, helping educators cope with professional stressors. These online interactions often compensate for the deficiencies in formal institutional support, suggesting that integrating such platforms into broader teacher support strategies could be beneficial. The study concludes that addressing teacher burnout requires a holistic approach, incorporating both organizational reforms and the utilization of digital platforms for peer support. The insights gathered underscore the importance of community and social exchange in mitigating burnout, pointing to potential pathways for improving teacher retention and well-being
MODULATING T CELL PHENOTYPE AND EXPANSION VIA CHANGING CO-STIMULTORY MOLECULE RATIO ON A FUNCTIONAL HYDROGEL
No full textAdoptive T cell therapy relies on ex vivo expansion of therapeutic T cells. During this ex vivo expansion, T cells often get exhausted and lose clinically favorable expression phenotypes. To address this concern, our lab has developed a biomimetic hydrogel functionalized with anti-CD3 antibodies, anti-CD28 antibodies, and interleukin-2 (IL-2) that inducessignificant T cell expansion and activation in vitro and in vivo.1 We have further shown that co-stimulating T cells by functionalizing multiple ligands, in addition to anti-CD28 antibodies, modulates cell phenotype and function. However, the relative importance of each co-stimulatory molecule in driving T cell outcomes remains unclear. To address this gap, we conjugated various ratios of co-stimulatory molecules to the hydrogel and investigated the effects of varying the ratio of these co-stimulatory molecules on T cell memory subset commitment and functional phenotypes, including how the synergistic impact of the combinations could be enhanced. Hydrogels conjugated with higher amounts of anti-CD28 co-stimulatory antibodies, specifically at a 4:1 ratio of anti-CD28 to other Signal 2A antibodies, resulted in a greater proportion of effector memory T cells. Conversely, hydrogels with higher levels of Signal 2A antibodies, particularly at a 1:3 ratio, promoted the expansion of central memory T cells. We also investigated whether presenting multiple antibodies on the same hydrogel versus separating the signals across different hydrogels would lead to distinct T cell outcomes. We observed that separating the molecules across multiple hydrogels led to significantly higher T cell expansion, while no major differences in functionality (as assessed by cytokine studies) were observed between the two configurations. Overall, our work uncovered new mechanistic insights into hydrogel-based stimulation of T cells that will have important implications for improving the performance of cellular therapies
Enhancing visual signal fidelity in a mouse model of retinitis pigmentosa
Get PDFMany inherited retina degeneration diseases target photoreceptors, which transduce light into a neural signal that is processed by the downstream visual system. When photoreceptors degenerate, physiological and morphological changes to retinal synapses and circuitry reduce sensitivity and increase noise, resulting in degrading visual signal fidelity. Retinitis Pigmentosa (RP) is the leading cause of inherited retinal degeneration and affects 1 in 4000 people worldwide across all ages. RP causes photoreceptors to degenerate and no longer encode and transmit visual information to the retina circuitry, gradually depriving patients of their eyesight. Here, we demonstrate that the rise of pathological noise during degeneration is a consequence of dysfunctional tonic synaptic transmission in the first synapse of the retinal circuitry, between photoreceptors and bipolar cells. We then pharmacologically targeted this synapse in an effort to ameliorate circuit noise without sacrificing visual sensitivity. We tested a novel strategy to partially replace the neurotransmitter lost when photoreceptors die with an agonist of receptors that ON bipolars cells use to detect glutamate released from photoreceptors. In rd10 mice, which express a photoreceptor mutation that causes RP, we found that a low dose of the mGluR6 agonist L-2-amino-4-phosphonobutyric acid (L-AP4) reduced pathological noise induced by photoreceptor degeneration. After making in vivo electroretinogram recordings in rd10 mice to characterize the developmental time course of visual signal degeneration, we examined effects of L-AP4 on sensitivity and circuit noise by recording in vitro light-evoked responses from individual retinal ganglion cells (RGCs). L-AP4 decreased circuit noise evident in RGC recordings without significantly reducing response amplitudes, an effect that persisted over the entire time course of rod photoreceptor degeneration. Subsequent in vitro recordings from rod bipolar cells (RBCs) showed that RBCs are more depolarized in rd10 retinas, likely contributing to downstream circuit noise and reduced synaptic gain, both of which appear to be ameliorated by hyperpolarizing RBCs with L-AP4. Although there are promising clinical trials investigating gene therapies for RP vision rescue, the genetically variable cause of RP is a major obstacle and reason for the lack of specific treatments available. However, this “catch-all” method using low-concentration L-AP4 to stabilize the inner retina circuitry even as photoreceptors die, regardless of photoreceptor mutation, could potentially reduce pathological circuit remodeling and preserve the efficacy of therapies designed to restore vision
CAREER PIPELINES TOWARD PROGRAM LEADERSHIP POSITIONS: EXPLORING ACCESS POINTS THAT MAY INCREASE DIVERSITY WITHIN THE FIELD OF GENETIC COUNSELING
Get PDFIn 2020, the National Human Genome Research Institute (NHGRI) published their strategic vision for the future of human genomics, identifying a goal of working toward a more diverse workforce at the forefront of their plan. Four years later, there is still a significant amount of work to be done in achieving the goal of diversifying the profession, requiring attention to many facets of the problem including barriers to diversity in genetic counseling graduate program leadership. Improving diversity among graduate students is directly aligned with NHGRI’s strategic vision, and there have been significant studies reporting the influence that a diverse team of program leadership and faculty has on bringing in new cohorts of graduate students from different backgrounds. However, despite their recognized influence on the future demographic of genetic counselors, assessment of genetic counselor career trajectories toward leadership positions has not yet been studied. Our study aims to shed light on the various carrier paths genetic counselors have taken in becoming members of program leadership, identifying common themes, experiences, and influential factors facilitating those carrier trajectories toward leadership positions. This study fits in nicely with the field’s overall goal of improving diversity by focusing on a population that has a significant influential power on the future direction of genetic counseling, while also filling in a gap in the current literature.
For this study, a survey collecting demographic information was sent to all current members of program leadership across the current 56 accredited genetic counseling programs in the United States. Additionally, 26 interviews were conducted with a purposefully selected sub-set of participants to explore career narratives of genetic counselors currently in program leadership positions. These interviews produced three types of career pipelines leading to program leadership, along with an analysis of facilitating factors identified in participant career trajectories. Following identification of those career advancing facilitators, a discussion on how these findings can be used to strategically improve the diversity among future members of program leadership and, by extension, the future of the field as a whole concludes this study
CATALYTIC UPCYCLING OF WASTE PLASTIC TO AROMATIC CHEMICALS
No full textThe accumulation of plastic waste represents a grand challenge as societies transition to sustainable and circular economies. Most plastic waste is not recycled and is either landfilled or incinerated. Much plastic waste also accumulates in the environment where it slowly degrades, contributes to general pollution, and disrupts ecosystems. Current recycling methods, i.e. mechanical recycling, are insufficient for maintaining plastic circularity for several reasons, including technical challenges, such as material property degradation, and economic challenges, such as the low price of oil compared to recycled plastic. At the same time, plastic waste is a carbon-rich material and could be used as a potential feedstock in the chemicals and manufacturing industries. For decades, researchers have studied methods for (thermo)chemically recycling plastics to lower molecular weight hydrocarbons and their derivatives, including waxes, lubricants, olefins, gasoline, and diesel, among others. Many of these (thermo)chemical processes result in low-value products or suffer from poor selectivity, yield, or scalability.
This dissertation presents systematic studies of virgin and waste plastic conversion to aromatic compounds, namely benzene, toluene, and xylenes (BTX) in a continuous fixed bed, two-stage, heterogeneous catalytic reaction system. We demonstrate that all types of plastic studied here can be fully converted with high selectivity towards BTX. In addition, kinetic studies are also discussed in order to elucidate the reaction mechanisms of the pyrolysis and catalytic reactions. Characterizations of the polymer feedstocks, pyrolysis wax intermediate, and catalyst are presented. Operational learnings in relation to general system operation, plastic pretreatment, plastic solid feeding, and others are discussed. These experimental studies are supplemented with technoeconomic analysis (TEA) and life cycle assessment (LCA) to assess the economic viability and carbon emissions of the proposed process at various industrial scales
MULTI-OMIC ANALYSIS OF MOLECULAR SIGNATURES ASSOCIATED WITH BREAST CANCER PROGRESSION IN PRECANCEROUS BREAST TISSUE
Get PDFBreast cancer is the most frequently diagnosed cancer in American women and ranks as the second leading cause of death among them. Identifying women at risk of developing invasive breast cancer (IBC) is crucial for reducing mortality from this disease. The transition to IBC involves progressive changes in the structure and composition of breast tissue; however, these morphological changes do not always lead to cancer development. Specifically, benign breast disease (BBD) and ductal carcinoma in situ (DCIS) are two breast conditions associated with an increased risk of IBC, yet they do not always result in breast cancer development. The challenge lies in determining who would benefit from intensive surveillance and treatment, and who may not need such interventions.
In this thesis, I tested the hypothesis that molecular changes can be detected in proliferative breast lesions prior to the development of IBC. Through two nested case-control studies, we investigated molecular signatures potentially linked to the progression of IBC. The first study focused on patients diagnosed with BBD who subsequently developed IBC, along with their corresponding tumors, and a matched cohort of BBD patients who did not progress to IBC. In this study, we examined genome-wide methylation patterns across these three cohorts.
The second study focused on patients diagnosed with DCIS who later progressed to IBC, compared to a control cohort of DCIS patients who did not progress to IBC. In this study, we employed a multi-omic approach, gathering three levels of genomic data: DNA methylation, gene expression, and DNA copy number variation. The gene expression data enabled us to assign PAM50 subtypes to each sample, while copy number variation was derived from methylation data using lab developed software (EpiCopy).
Both studies revealed molecular differences between progressors and non-progressors. However, the interpretation of these findings is complicated by the heterogeneity of breast diseases and the small sample sizes within subsets. Despite these challenges, both studies present promising avenues for future research
Spectral and Temporal Dynamics of Brief Arousals During Sleep
No full textBrief arousals (BAs) are short, wake-like events which occur during healthy non-rapid eye movement sleep (NREMS) and typically last under 16 seconds in rodents. BAs are characterized by sudden shifts in EEG frequency accompanied by an increase in EMG activity. These transient spontaneous arousals are theorized to play a role in sleep reversibility, threat vigilance, and memory consolidation, often occurring in response to external stimuli such as noises and light as well as under conditions of stress, anxiety, pain, and other disorders. Understanding the causes of BAs can help improve sleep quality and overall health. However, studying BAs is challenging due to their heterogeneity, presence in both health and in disease, and duration variability.
In this thesis work, we investigated how the spectral and temporal characteristics of BAs in mice evolve in baseline (BSL) conditions and after spared nerve injury (SNI) during the light and dark periods. 24-h EEG and EMG data were collected from eight mice.
We created a logistic regression model which predicted, given an arousal event, whether NREMS transitions to a BA or a longer wake event with a 0.73 AUC for light period BSL, 0.77 AUC for light period SNI, 0.76 AUC for dark period BSL, and 0.79 AUC for dark period SNI using EEG from 30 seconds prior to the event. For cases where NREMS was followed by a BA, we found correlations between BA durations and spectral features preceding and during the arousal (R2 = 0.59 for light period BSL, 0.57 for light period SNI, 0.54 for dark period BSL, 0.51 for dark period SNI). Furthermore, we identified a peak in inter-BA intervals in the SNI condition at 5 seconds which was absent in the BSL condition during the light period.
This work highlights and characterizes the diversity of BA characteristics across light and dark periods in mice with and without neuropathic pain
NEUROPHOTONIC INTERFACE FOR BIO-ELECTRONIC AND BIO-MAGNETIC RECORDING
No full textIn recent decades, in-vivo neuron activity recording has been developed to explore the complex behavior of brain function. This has been achieved with the help of deep neural multi-site probe arrays, integrated stimuli-recording probing techniques, and non-invasive skull electrodes for bio-electronic recording. Additionally, whole-brain imaging technologies such as magnetic resonance imaging, magnetoencephalography, and electroencephalography have been utilized for bio-magnetic recording. Current improvements aim to record a large amount of neuron activity in the long term as well as minimize cortex inflammation if extra implantation surgery is needed. The commonly used micro-electrode array turns out to be the bottleneck in developing the next generation of electrodes due to its weak stiffness, signal crosstalk among traces, large device sensing site, and heat generated in small volumes. Moreover, MEG and similar equipment suffered from time-spatial resolutions, expensive maintenance costs, and limited accessibility. Recently, in order to tackle the problem, researchers have devoted themselves to finding a feasible way to address the above issues on the photonics platform. In this thesis, we come up with a novel design that utilizes the mature fabrication technology from the CMOS industry and the ongoing developments of photonics, demonstrating the capability of photonics in benefiting biology signal detection and activity analysis.
The first experimental chapter focuses on the design, fabrication, and in-vivo testing of silicon-based photonics sensors for accurate electrocardiogram(EKG) recording. The device, based on carrier injection, accomplishes sub-milli volts of electronic signal measurements. The experimental demonstration of using a carrier injection mode device provides another track for the development of measuring bio-electronic signals and examining corresponding brain or heart activities.
The second experimental chapter demonstrates the feasibility of sensing a magnetic field with a hybrid silicon nitride device coated by a magnetic optical polymer. The experimental verification paves the way for building a CMOS-compatible photonic magnetometer that is small in footprint, stable at room temperature, and ultra-sensitive.
The content described in this thesis addresses the issue of implementing photonics in bio-electronic and magnetic recording, enabling future developments in the neuron-photonics interface and ultimately leading to a complete revolution of the current strategy for identifying biological features