48670 research outputs found
Sort by
Visualizing Genetics: Quantifying Gene Expression through Genetic and Epigenetic Modifications
No full textPrecise control over gene expression is essential for proper development of an organism. However, genes are regulated in a variety of ways, the mechanisms of which are not well understood. This thesis explores gene regulation at two different levels: genetic control and epigenetic control. Genetic control focuses on components of the genome such as enhancers, promoters, gene composition, and gene length. Each of these facets can potentially affect either the timing, amount, or location of gene expression. Epigenetic control looks at how different DNA markers, which do not modify the DNA sequence, are able to manipulate gene production. Using quantitative live-imaging in Drosophila embryos and immunofluorescence staining in human fibroblasts, we are able to capture and analyze the impact each of these factors have on gene expression. In genetic control, we see that enhancers play a dominant role in controlling expression, specifically through the individual transcription factor binding sites within an enhancer. Based on the binding affinity and type of binding site, the enhancer can influence when genes are active as well as the amount of mRNA produced. The level of mRNA production is also influenced by the rate of RNA polymerase II (Pol II) elongation. The speed that Pol II moves along a gene body can impact how much mRNA is made within a certain developmental time period. We find that not only do enhancers play a role in controlling Pol II elongation rate, but the composition of the gene itself also contributes to modulating the rate of elongation. In epigenetic control, there are many different markers interacting with DNA. Here we identified H3K9me3 as a key epigenetic modification that controls DNA compaction, and thus can silence a large number of genes. As a result, specific epigenetic markers can ultimately control gene expression and dramatically impact a cell’s ability to reprogram itself. Slight changes in gene regulatory mechanisms can cause extreme changes in gene expression and ensuring that organisms develop properly is contingent upon having a better understanding of how different factors influence expression
Melanin Synthesis Intermediates Inhibit Melanomagenesis
No full textMelanocytes are specialized cells responsible for producing and providing melanin pigment to mammalian skin and hair. The variation in human skin pigmentation has evolved through natural selection in response to ultraviolet radiation (UVR). For decades, the connection between UVR exposure and the incidence of melanoma has been appreciated. However, populations living in the same geographic region, and thereby exposed to similar amounts of UVR, exhibit large variations in the lifetime risk of cutaneous melanoma. By understanding the mechanisms by which melanoma differentially affects diverse populations, we may reveal new therapeutic approaches. In the United States, the risk of melanoma is 30 times higher in people with lightly pigmented skin versus darkly pigmented skin. Researchers have long suggested that this large fold difference cannot simply be explained by the UVR-protective effect of melanin pigment, but the specific mechanisms have remained unknown. During this thesis work, we show that lightly pigmented melanocytes (LMCs) have cell-intrinsic differences that cause them to be more susceptible to melanomagenesis than darkly pigmented melanocytes (DMCs). We go on to show that these differences result from dihydroxyphenylalanine (DOPA), a melanin synthesis intermediate, which is synthesized and secreted from DMCs at higher levels than LMCs. DOPA’s anti-proliferative effect is independent of melanin synthesis and exogenous DOPA can recapitulate DMC phenotypes in LMCs. Using complimentary pharmacologic and genetic in vivo screens, we found that DOPA limits melanocyte and melanoma cell proliferation through inhibition of muscarinic acetylcholine receptor M1 (CHRM1) signaling. We then determined that DOPA’s antagonism of CHRM1 converged on two transcription factors and major cell cycle regulators, FOXM1 and c-Myc. Systemic treatment of melanoma-bearing mice with specific novel FOXM1 inhibitors was tolerated and had dramatic anti-tumor effects, with some mice exhibiting complete and sustained tumor clearance. In conclusion, this work identified the first UV-independent cell-intrinsic mechanism by which highly pigmented melanocytes are protected against melanomagenesis, first to propose that DOPA regulates CHRM1 signaling, and discovered two novel therapeutic targets for melanoma, CHRM1 and FOXM1
Contingent Citizenship: Muslims in America
No full textHow do American Muslims experience political incorporation? What are the consequences of political incorporation for Muslim perceptions of American citizenship and (dis)engagement in civic and political life? In this study, these questions are investigated in-depth across three different contexts and with different communities of Muslims in the United States. Drawing on ethnographic interviews and participant observation with Muslim communities, I find that the status of Muslims as incorporated, excluded, or somewhere along the continuum between, is a function of the interplay between the intersectionality of American Muslim group identity and their local political context. As I will argue throughout this study, the membership or status of Muslims varies locally depending on the set of actors and institutions that receive Muslims, the dimensions of American Muslim group identity which are brought to the forefront by these actors and institutions, and the coalitions that emerge as Muslims are politically socialized
Sex-Specific Regulation of Promoter Bivalency in the Mouse Striatum
No full textCocaine addiction is a major public health concern with no FDA approved treatment options. Understanding the epigenetic mechanisms that mediate relapse may lead to the development of new therapies. Recent evidence shows that changes to histone posttranslational modifications (HPTMs) occur and persist through drug use and abstinence periods. These epigenetic changes alter gene expression, yet the underlying molecular mechanisms remain elusive. Bivalent chromatin domains encompass both gene activating and repressive HPTMs, such as H3K4me3 and H3K27me3, respectively. Expression of the immediate early gene, Nr4a1, is activated immediately following mouse cocaine exposure, and then returns to baseline during abstinence. We hypothesized that Nr4a1 chromatin bivalency regulates transient cocaine-induced Nr4a1 activation, in male and female mouse striatum. Accordingly, we measured mRNA and bivalent HPTMs using sequential chromatin immunoprecipitation in multiple brain regions of the reward pathway following cocaine. We found that cocaine induces a sex-specific effect on Nr4a1 promoter bivalency and corresponding activation in the striatum. As transcription differs between the two major cell types of the striatum, we then sought to investigate bivalency at the cell-type specific level. We established transgenic mouse lines that express an affinity tagged nuclear receptor (GFP-SUN1 fusion) in dopamine D1 receptor- or adenosine A2A receptor-containing cell types. We were able to identify HPTMs at specific gene promoters in these specific cell-types by our hybrid protocol, ICuRuS, that combines INTACT (Isolation of Nuclei Tagged in Specific Cell Types), CUT&RUN (Cleavage Under targets and Release Using Nuclease), and next generation Sequencing. We present data on the protocol optimization as evidence that we successfully established a stream-lined method of cell-type specific quantification of bivalent chromatin. Future studies will apply this methodology to the broader hypothesis that bivalent chromatin plays a prominent role inducing HPTMs and altering persistent gene expression across drug use and abstinence. Taken together, this dissertation presents a novel hybrid method to broadly uncover cell-type specific bivalent promoters and provides mechanistic insights into the sex- and region-specific cocaine regulation of Nr4a1
Designing Hyaluronic Acid Granular Hydrogels for Biomaterials Applications
No full textIn recent years, there has been growing interest in the use of granular hydrogels as biomaterials for biomedical applications. Granular hydrogels consist of hydrogel microparticles (i.e., “microgels”) that are tens to hundreds of microns in diameter and packed into a jammed state. This thesis investigates the following central question: how do we best fabricate and design granular hydrogels with modular properties for biomedical applications? To do so, granular hydrogels are made from fragmented hyaluronic acid (HA) microgels and explored for applications in injectable tissue repair, extrusion printing, and cell culture. The material properties are explored with respect to changes in particle design and assembly properties (i.e., microgel shape and size distribution, degree of jamming, intra- and inter-particle crosslinking chemistry, addition of interstitial phase hydrogels) and thoroughly characterized. First, the properties of HA granular hydrogels were characterized as a function of microgel fabrication method (i.e., microfluidic devices, batch emulsions, and mechanical fragmentation by extrusion) towards informed biomaterial design. Heterogeneous and jagged microgels fabricated from extrusion fragmentation yields granular hydrogels with enhanced mechanical moduli and structural integrity compared to spherical microgels fabrications from emulsion-based techniques. Next, the mechanical properties of granular composites were tailored by varying microgel and interstitial matrix compositions and moduli. Granular hydrogel composites consist of microgels are embedded in a crosslinked interstitial matrix, which significantly improves mechanical moduli compared to granular hydrogels without an interstitial matrix. While compressive moduli increased with increasing microgel modulus and interstitial matrix modulus, it was determined that failure properties (i.e., strain and stress) increased by combining softer microgels with stiffer interstitial matrices. Types of crosslinkers (i.e., covalent v. guest-host) as well as degradability were varied in each phase to further understand material properties towards informed biomaterials deign for future applications. Next, injectable and adhesive granular hydrogels with dynamic-covalent interparticle crosslinking were investigated for 3D printing and cell culture applications. Introducing dynamic covalent hydrazone bonds between microgels resulted in significant increases in structural stability while maintaining injectability. Adhesive granular hydrogels were used as a 3D printing ink, where printed structures were immediately stable upon deposition without the need for post-process steps. Further, adhesive granular hydrogels allow for cell invasion through an in vitro spheroid outgrowth assay. This work demonstrates the use of dynamic covalent inter-particle crosslinking to enhance injectable granular hydrogels. Lastly, injectable radiopaque granular hydrogel was fabricated for intervertebral disc repair. Zirconium oxide (ZrO2) nanoparticles were encapsulated into microgels to introduce radiopacity, enabling direct visualization of the hydrogel using clinically relevant imaging technologies (i.e., x-ray and CT scan). Radiopaque granular hydrogels restored healthy disc mechanics in a degenerative disc rabbit model ex vivo. As a proof-of-concept, the radiopaque granular hydrogel was directly visualized following percutaneous intradiscal delivery in a degenerated goat disc in vivo. Overall, this study demonstrates the great potential of injectable radiopaque granular hydrogels for degenerative disc disease treatment. The use of granular hydrogels for biomedical applications grows each year, and the work in this thesis can help guide material development to advance granular hydrogel biomaterials
Topics in Statistical Machine Learning
No full textModern statistical machine learning combines statistics with the computational sciences including computer science and optimization. The research in statistical machine learning facilitates the development of fields such as medicine, signal processing, bioinformatics, artificial intelligence and operation research and these fields provide practical problems that motivate statistical machine learning study. Nowadays, more and more opportunities can be found in statistical machine learning driven by these applied problems. This thesis aims to address the following three problems in statistical machine learning: In the first part, we study stochastic continuum-armed bandits with additive models. A near optimal algorithm is proposed and the minimax rate of regret is established. The results show an interesting phenomenon: the optimal regret is independent of the dimension if sparsity assumption is made, which highlights the difference between high-dimensional bandits and high-dimensional estimation. We also study the adaptivity issue of this problem and show it is possible to adapt to the sparsity but impossible to adapt to the smoothness. We then develop a new algorithm that can achieve near optimal regret adaptively under an additional assumption. In the second part, we study the problem of transfer learning for nonparametric regression. A near optimal algorithm is also established and the minimax optimal rate of risk is identified. We further propose a data-driven algorithm and show it simultaneously attains the optimal rate over a large collection of parameter spaces. Finally we extend this problem to the case where multiple source domains are considered. In the third part of this thesis, we study the optimal treatment rules with an instrumental variable in causal inference. We first propose a general framework for estimation of optimal individualized treatment rules with a valid instrumental variable. Under this framework, we then define a novel notion of optimality called IV-optimality for treatment rules. Finally we propose an estimator of an IV-optimal rule and prove theoretical guarantees
The Molecular Basis of Human Adaptation in Sub-Saharan Africa
No full textGenomics has provided unprecedented opportunities for understanding the genetic architecture and evolutionary history of human traits. Unfortunately, human genomics has been persistently biased toward the study of populations of European-derived ancestry, limiting our understanding. The study of diverse and non-European populations can identify novel genetic loci underlying trait variation, give a clearer picture of human demographic history inside and outside of Africa, and identify how humans have adapted to their environment. In this thesis I present work from two studies; the first aims to identify genetic variants, genes, and molecular pathways contributing to skin pigmentation variation in sub-Saharan Africa, as well as the evolutionary history of this genetic variation. To this end, I combine genome-wide association data with functional genomic data and tools, identifying candidate causal genes and variants, along with their predicted regulatory effects. With the addition of scans of selection and global frequency data, we identify the genes targeted by selection on pigmentation, and how that variation is shared across populations. In the second study, I combine genotype data from a cohort of East Africans with whole blood gene expression data to identify genetic variants underlying gene regulatory variation. I compare these associations with findings from other studies to evaluate the extent of sharing of gene regulatory architecture across populations and compare ability to fine-map QTL signals, finding overall strong replication of QTLs and improved fine-mapping in Africans compared with European-Americans. Finally, the integration of QTL data with signatures of selection identifies genes targeted by selection in specific populations, including a gene which may underly skin pigmentation variation and is under selection in Nilo-Saharan speaking populations
High Throughput Microfluidics for Ultrasensitive Blood-Based Diagnostics
No full textUltrasensitive diagnostic assays have made it to the bench in recent years and helped improve the clinical landscape in many diseases. These assays can target nucleic acids, proteins, extracellular vesicles or whole cells down to the single particle, thus allowing for early detection and continuous monitoring of cancers and infectious diseases. However, the two primary limitations to these technologies are that 1. they require complicated and lossy sample processing steps to remove biological background from and 2. their throughput is limited to single-particle analysis to maximize sensitivity. As a result, we have developed two platform technologies that can be used with blood-based liquid biopsies for detection of rare targets at a much higher throughput than is currently being implemented. We first demonstrate the ability of graphene Hall sensors to detect magnetic particles directly from blood. Using a process compatible with CMOS technology used in integrated circuits, we demonstrate long-term stability and high sensitivity in a hybrid microfluidic-microelectronic chip for eventual application with rare cell detection from whole blood. We then developed a digital droplet ELISA platform for the ultrasensitive detection of p24, a highly-conserved HIV protein, that can improve the limit of detection by greater than 2 orders of magnitude. Using a cellphone- based imaging platform for high throughput droplet detection, this platform could easily be implemented for continuous at-home monitoring of p24 in HIV patients on antiretroviral therapies at risk of virological rebound. Together, these projects lay the groundwork for ultrasensitive diagnostic tests that can be accomplished with minimal processing sample and maximum throughput. Although we present two applications of these platforms, they can easily be extended for detection of multiple biomarker modalities across clinical applications. Additionally, by combining advancements in microfluidics and microelectronics into fully integrated assays, these platforms can be used as point-of-care tools for optimal disease management in a personalized manner
Accuracy of Intra-oral Scanner Impression for Different Post-Space Lengths
No full textPurpose: To compare the accuracy of the intra-oral scanner for different post space lengths 6, 8, and 10 mm and compare it with conventional impressions using three-dimensional software.
Materials and methods: A total of forty-five root typodont teeth (Maxillary central incisors) were selected. Root canal treatment and tooth preparation for the crown were performed on all teeth. Post space preparation of 6, 8, and 10 mm were created using prefabricated fiber post drill to standardize post space width and length. Root canal impressions were performed on all teeth using polyvinyl siloxane impression material. Each impression was then three-dimensionally scanned using an extraoral lab scanner to be used as a control. A direct scan was obtained using the chair-side scanner for each group. The file was exported to perform digital volume measurement using Geomagic control software to determine accuracy.
Results:Mean overall post-length accuracy was 75, 95, and 144 mm for groups 6, 8, and 10 mm post-depth, respectively. The accuracy of the digital intra-oral scanner compared to PVS impression groups was statistically not significant in the coronal, middle, and apical third (P= \u3e0.05) except in the apical third for the 10 mm post depth (P=
Conclusion: Direct chairside scanning for post space has the same comparable accuracy as those using the gold standard direct conventional impression technique
A Just and True Return: A Dataset of Pennsylvania\u27s Surviving County Slave Registries
No full textA Just and True Return (JATR) contains information about more than 6,300 Black people and their enslavers principally taken from extant registries from fifteen Pennsylvania counties: Adams, Allegheny, Bedford, Berks, Bucks, Centre, Chester, Cumberland, Dauphin, Delaware, Fayette, Lancaster, Northampton, Washington, and Westmoreland. It also includes a handful of records from four counties—Crawford, Franklin, Philadelphia, and York—whose registries have not been located, but which can be partially reconstructed from a variety of other sources. Pennsylvania\u27s 1780 gradual abolition law required enslavers to register with their county clerk any people they wished to continue holding in lifetime slavery. A 1788 law required that they do the same for any children they wished to hold in twenty-eight-year term slavery. Complete entries provide the name, age or birthday, race, and sex of enslaved people; the name, place of residence, and occupation of their enslavers; and the registration date. Slightly less than two-thirds of the entries describe people whom enslavers held in lifetime slavery, whereas more than one-third describe children they held in term slavery. An ongoing project, JATR is the first effort to compile all surviving registration data in a single location and contributes to our understanding of slavery’s survival in the northern United States during the early republic