Browsing by Subject "Infectious disease"

Unprecedented amount of data has been collected in diverse fields such as social network, infectious disease and political science in this information explosive era. The high dimensional, complex and heterogeneous data imposes tremendous challenges on traditional statistical models. Bayesian nonparametric methods address these challenges by providing models that can fit the data with growing complexity. In this thesis, we design novel Bayesian nonparametric models on dataset from three different fields, hyperspectral images analysis, infectious disease and voting behaviors.

First, we consider analysis of noisy and incomplete hyperspectral imagery, with the objective of removing the noise and inferring the missing data. The noise statistics may be wavelength-dependent, and the fraction of data missing (at random) may be substantial, including potentially entire bands, offering the potential to significantly reduce the quantity of data that need be measured. We achieve this objective by employing Bayesian dictionary learning model, considering two distinct means of imposing sparse dictionary usage and drawing the dictionary elements from a Gaussian process prior, imposing structure on the wavelength dependence of the dictionary elements.

Second, a Bayesian statistical model is developed for analysis of the time-evolving properties of infectious disease, with a particular focus on viruses. The model employs a latent semi-Markovian state process, and the state-transition statistics are driven by three terms: ($i$) a general time-evolving trend of the overall population, ($ii$) a semi-periodic term that accounts for effects caused by the days of the week, and ($iii$) a regression term that relates the probability of infection to covariates (here, specifically, to the Google Flu Trends data).

Third, extensive information on 3 million randomly sampled United States citizens is used to construct a statistical model of constituent preferences for each U.S. congressional district. This model is linked to the legislative voting record of the legislator from each district, yielding an integrated model for constituency data, legislative roll-call votes, and the text of the legislation. The model is used to examine the extent to which legislators' voting records are aligned with constituent preferences, and the implications of that alignment (or lack thereof) on subsequent election outcomes. The analysis is based on a Bayesian nonparametric formalism, with fast inference via a stochastic variational Bayesian analysis.

Zoonotic disease poses one of the greatest threats to both human and animal health in our world today. Recent pandemics such as Ebola, SARS-CoV2, and highly pathogenic Avian Influenza have shown the devastating consequences of infectious disease on both human and animal populations. Few studies have characterized the risk of interspecies pathogen transmission with one of our closest living relatives, Pan paniscus. This endangered species of great apes inhabits only rainforests of the Congo River Basin. Due to human encroachment, bushmeat trade, illegal pet trade, and deforestation, only 15,000-20,000 wild bonobos remain. The Lola Ya Bonobo Sanctuary (LYB) serves as the world’s only haven for orphaned and injured bonobos and presents a valuable opportunity to study the epidemiology of respiratory pathogens in a habituated population. This study aimed to characterize the risk factors for transmission of respiratory illness between bonobos and humans at LYB. Between 2014-2017 a cohort of 77 bonobos and 44 human staff were observed for signs of clinical illness and had their upper respiratory tract routinely sampled. We used a multiplex nucleic acid amplification assay to detect the presence of 18 viral and 3 bacterial respiratory pathogens. Of 282 total bonobo testing events, 93 (33%) returned positive results, primarily RSV-A (n = 41, 44.1%) and Rhinovirus/Enterovirus (n = 37, 39.8%). Although the point of entry into LYB could not be determined, RSV-A initially appeared in the infant and juvenile enclosures and spread serially through the older bonobo enclosures. Rhinovirus/Enterovirus appeared in diffuse clusters throughout the sanctuary. Although new bonobos are subject to a prolonged quarantine, current policy allows for relocation of bonobos between enclosures and frequent and sustained contact with human staff. Active surveillance for respiratory and other pathogens and additional infection control measures may benefit bonobo and human health.

Understanding the complex interactions between humans and their pathogens is key to the development of effective therapeutic strategies for infectious diseases. One approach to gain insight into host-pathogen interactions is to leverage natural human genetic variation. Traditionally, researchers have employed clinical GWAS (genome-wide association studies) of infected individuals to identify genetic variants that confer susceptibility to infection phenotypes. However, standard clinical GWAS approaches are hampered by issues with sampling, variation in exposure, and difficulty obtaining appropriately matched controls. In this thesis, I have leveraged cellular and molecular GWAS of lymphoblastoid cell lines (LCLs) to uncover mechanisms of immune suppression by Chlamydia trachomatis and identify novel regulators of influenza infection.Previously, our lab developed Hi-throughput Human in vitrO Susceptibility Testing (Hi-HOST) to connect human genetic variation to infectious disease phenotypes measured by flow cytometry and immunoassays from cellular infection of LCLs. Applying Hi-HOST to Chlamydia trachomatis, an obligate intracellular bacterium, revealed a genome-wide significant association between rs2869462 and levels of a pro-inflammatory chemokine, CXCL10, measured in assay supernatants before and after infection. Curiously, we noticed wide variation in induction of CXCL10 that was not associated with rs2869462. Leveraging flow cytometric measurements of infected cells in a multivariate linear model revealed that the most highly infected LCLs showed a high degree of suppression of CXCL10. This indicated to us that C. trachomatis may be actively suppressing CXCL10 induction. We experimentally identified chlamydial protease-like activity factor (CPAF) as responsible for suppression of CXCL10. Applying our multivariate modeling to a panel of 17 other cytokines revealed a similar signature of suppression for RANTES. However, this phenotype was not mediated by CPAF, indicating some degree of specificity of CPAF activity. To further refine Hi-HOST with higher resolution phenotypes and integrated eQTL (expression quantitative trait loci) analyses all in a single infection, we developed single-cell Hi-HOST (scHi-HOST). scHi-HOST leverages single-cell RNA-sequencing of pooled LCLs to simultaneously identify alleles associated with gene expression and susceptibility to influenza A virus (IAV). scHi-HOST identified a common missense variant in ERAP1, rs27895, as associated with viral burden in LCLs. I confirmed this association experimentally using RNAi, overexpression and small molecule inhibition of ERAP1 in vitro. Finally, we performed analysis of human flu challenge and found that volunteers with the risk allele of rs27895 had increased viral burden and worse symptoms over the course of their infection, indicating that our cellular findings may translate to human flu susceptibility as well. Finally, to identify strain-specific susceptibility alleles, I applied scHi-HOST to six diverse strains of IAV. Analyses of these data suggested that infection with CA09 (the strain responsible for the 2009 “Swine Flu” pandemic, A/California/04/2009), produced distinct infection phenotypes and a distinct set of associated genetic variants relative to other strains. I identified rs7144228, an eQTL for HSP90AA1, as significantly associated with CA09 infection, but not any other IAV strain. rs7144228 is specific to populations with African ancestry and contributes more broadly to population differences observed during IAV infection of LCLs. I also identified rs113816500, a SNP intronic to CTSH, as associated with all six strains of IAV, and therefore is a conserved host factor that influenza exploits to increase viral burden. This study suggests that susceptibility to infection is not only dependent on the genotype of the affected individual but is also dependent on the genetic background of the virus.

In this dissertation, I present a general statistical framework for phylodynamic inference that can be used to estimate epidemiological parameters and reconstruct disease dynamics from pathogen genealogies. This framework can be used to fit a broad class of epidemiological models, including nonlinear stochastic models, to genealogies by relating the population dynamics of a pathogen to its genealogy using coalescent theory. By combining Markov chain Monte Carlo and particle filtering methods, efficient Bayesian inference of all parameters and unobserved latent variables is possible even when analytical likelihood expressions are not available under the epidemiological model. Through extensive simulations, I show that this method can be used to reliably estimate epidemiological parameters of interest as well as reconstruct past disease dynamics from genealogies, or jointly from genealogies and other common sources of epidemiological data like time series. I then extend this basic framework to include different types of host population structure, including models with spatial structure, multiple-hosts or vectors, and different stages of infection. The later is demonstrated by using a multistage model of HIV infection to estimate stage-specific transmission rates and incidence from HIV sequence data collected in Detroit, Michigan. Finally, to demonstrate how the approach can be used more generally, I consider the case of dengue virus in southern Vietnam. I show how earlier phylodynamic inference methods fail to reliably reconstruct the dynamics of dengue observed in hospitalization data, but by deriving coalescent models that take into consideration ecological complexities like seasonality, vector dynamics and spatial structure, accurate dynamics can be reconstructed from genealogies. In sum, by extending phylodynamics to include more ecologically realistic and mechanistic models, this framework can provide more accurate estimates and give deeper insight into the processes driving infectious disease dynamics.

Linker for activation of T cells (LAT) is a transmembrane adaptor protein that is crucial in linking TCR engagement to downstream signaling events, such as calcium flux and Ras-MAPK pathway. Following TCR engagement, LAT is phosphorylated at its membrane-distal tyrosine residues, which mediates the binding of Grb2/Sos, PLC-1, and GADS/SLP-76 complexes. This multi-protein signaling complex initiates signaling cascades eventually leading to the activation of transcription factors that regulate the genes required for T cell proliferation and effector functions. The indispensable role of LAT in thymocyte development has been evidenced as LAT-deficient mice completely lack peripheral T cells. To study the function of LAT in mature T cells, our lab previously generated a conditional knock-in mouse line in which the lat gene can be deleted by Cre recombinase. Deletion of LAT in mature T cells revealed the critical role of LAT in T cell activation. Here, we used this inducible LAT deletion mouse line crossed with the OT-I transgenic mice to study the role of LAT in mature CD8 T cells.

To analyze the contribution of LAT in CD8 T cells during the course of pathogen infection, we infected mice with Listeria monocytogenes-expressing Ova to elicit activation of antigen-specific CD8 T cells, and then inducibly deleted LAT in these cells at different stages of infection under the control of tamoxifen treatment. We show that LAT is important for maintaining CD8 T cell expansion during the priming phase; however, it is not required for CD8 T cell contraction. In addition, memory CD8 T cell can persist in the absence of LAT, suggesting that LAT-signaling is not necessary for memory maintenance. Nonetheless, these LAT-deficient memory T cells were unable to proliferate or produce cytokines upon secondary infection. Moreover, LAT deficiency accelerates memory differentiation during the effector-to-memory transition, leading to a higher frequency of KLRG1lowIL-7RhighCD62Lhigh memory T cells. Together, these data demonstrate that, while it is dispensable for contraction and memory maintenance, LAT-signaling regulates CD8 T cell memory differentiation and is essential for the memory response against pathogens.

The fundamental activity of CD8 T cells is to elicit cytotoxicity toward target cells that express foreign antigens, and this is mediated through granule-dependent and Fas ligand-dependent mechanisms. The signaling events that regulate these processes remain unclear. We showed that LAT-deficient cytotoxic T cells (CTLs) failed to upregulate FasL and produce IFN- after engagement with target cells. Moreover, they displayed reduced granule-mediated killing. We further dissected the effect of the LAT deletion on each step of granule exocytosis. LAT-deficiency led to altered synapse formation, subsequently causing unstable T cell:APC conjugates. MTOC polarization and granule reorientation were also impaired by LAT-deficiency, leading to reduced granule delivery. Despite these defects, granule release was still observed in LAT-deficient CTLs due to residual calcium flux and PLC activity. This revealed an unexpected finding that CTL function is not entirely dependent on LAT. Collectively, these data indicate that the signaling circuits governing CTLs are programmed to adopt multiple pathways, allowing CTLs to effectively eliminate various pathogens during adaptive immune responses.

Introduction: Municipalities often struggle to build and maintain basic infrastructure for informal slums in urban cities for its most vulnerable populations. One impact of inadequate water and sanitation access is the creation of an environment that breeds water borne pathogens that are the agents of infectious disease. Escherichia coli is a common bacteria found in water, often as an indicator of fecal contamination in the water supply. This study looks at one of the most common diseases found in women that results from E. coli growth, urinary tract infection. Specifically, this study aims to examine and describe factors of water, sanitation, and hygiene that are associated with positive E. coli urine results among women. The study took place in Ahmedabad, Gujarat, one of India's wealthier cities, in which heavy investments have been made in improving slum settlements throughout the rapidly expanding city.

Method: This was a cross-sectional study of 250 women recruited from households in urban Ahmedabad from October to December 2017. To determine positive cases of E. coli urinary tract infection, urine samples were collected from each participant. A commercial laboratory performed the urine analyses using a culture method. The threshold for positive cases was 10,000 CFU/mL or greater for E. coli. To obtain information on the water, sanitation, and hygiene practices, each participant completed a structured survey that included questions on demographics, working environment, reproductive health, sanitation access, family relationships, public toilets and social customs.

Results: Of the 250 participants, 23 (9.2%) were above the 100,000 CFU/mL threshold for E. coli, and therefore defined as a positive case. There were 124 (49.6%) participants who attempted a treat method, such as over the counter medicine or home self-treatments, for feminine health in the last three month. There were three factors that significantly correlated with positive cases. The first was the location of the handwashing facility, which could be either inside or outside of the dwelling. The second factor was antibiotic use in the last three weeks. The third factor was a participant living in a home with a child under the age of 5 years old, who experienced diarrhea.

Conclusion: This study identified a higher point prevalence of positive E. coli urine cultures than what we would want or would have expected for a sample population that all had access to piped water and a toilet inside of the dwelling. There is evidence to suggest that hygiene management around water use has an impact on a woman's susceptibility for E. coli causing infections in the urinary tract. Because half of the participants sought a form of treatment over the last three months for feminine health, a longitudinal study that tracks these women over a three month period, could provide relevant information on the incidence of new infections as well as prolonged urinary tract infections, particularly since multi-drug resistant E. coli infections are on the rise.