Browsing by Author "Tegge, Allison"
Now showing 1 - 8 of 8
Results Per Page
Sort Options
- Bayesian Factor Models for Clustering and Spatiotemporal AnalysisShin, Hwasoo (Virginia Tech, 2024-05-28)Multivariate data is prevalent in modern applications, yet it often presents significant analytical challenges. Factor models can offer an effective tool to address issues associated with large-scale datasets. In this dissertation, we propose two novel Bayesian factors models. These models are designed to effectively reduce the dimensionality of the data, as the number of latent factors is typically much smaller than that of the observation vectors. Therefore, our proposed models can achieve substantial dimension reduction. Our first model is for spatiotemporal areal data. In this case, the region of interest is divided into subregions, and at each time point, there is one univariate observation per subregion. Our model writes the vector of observations at each time point in a factor model form as the product of a vector of factor loadings and a vector of common factors plus a vector of error. Our model assumes that the common factor evolves through time according to a dynamic linear model. To represent the spatial relationships among subregions, each column of the factor loadings matrix is assigned intrinsic conditional autoregressive (ICAR) priors. Therefore, we call our approach the Dynamic ICAR Spatiotemporal Factor Models (DIFM). Our second model, Bayesian Clustering Factor Model (BCFM) assumes latent factors and clusters are present in the data. We apply Gaussian mixture models on common factors to discover clusters. For both models, we develop MCMC to explore the posterior distribution of the parameters. To select the number of factors and, in the case of clustering methods, the number of clusters, we develop model selection criteria that utilize the Laplace-Metropolis estimator of the predictive density and BIC with integrated likelihood.
- Bayesian variable selection for linear mixed models when p is much larger than n with applications in genome wide association studiesWilliams, Jacob Robert Michael (Virginia Tech, 2023-06-05)Genome-wide association studies (GWAS) seek to identify single nucleotide polymorphisms (SNP) causing phenotypic responses in individuals. Commonly, GWAS analyses are done by using single marker association testing (SMA) which investigates the effect of a single SNP at a time and selects a candidate set of SNPs using a strict multiple correction penalty. As SNPs are not independent but instead strongly correlated, SMA methods lead to such high false discovery rates (FDR) that the results are difficult to use by wet lab scientists. To address this, this dissertation proposes three different novel Bayesian methods: BICOSS, BGWAS, and IEB. From a Bayesian modeling point of view, SNP search can be seen as a variable selection problem in linear mixed models (LMMs) where $p$ is much larger than $n$. To deal with the $p>>n$ issue, our three proposed methods use novel Bayesian approaches based on two steps: a screening step and a model selection step. To control false discoveries, we link the screening and model selection steps through a common probability of a null SNP. To deal with model selection, we propose novel priors that are extensions for LMMs of nonlocal priors, Zellner-g prior, unit Information prior, and Zellner-Siow prior. For each method, extensive simulation studies and case studies show that these methods improve the recall of true causal SNPs and, more importantly, drastically decrease FDR. Because our Bayesian methods provide more focused and precise results, they may speed up discovery of important SNPs and significantly contribute to scientific progress in the areas of biology, agricultural productivity, and human health.
- Delay Discounting as a Potential Therapeutic Target for Weight Loss in Breast Cancer SurvivorsSukumar, Jasmine S.; Vaughn, Jennifer E.; Tegge, Allison; Sardesai, Sagar; Lustberg, Maryam; Stein, Jeffrey (MDPI, 2022-03)Simple Summary Obesity is a rising health epidemic in breast cancer survivors and associated with multiple negative health sequalae and increased mortality. Delay Discounting (DD) is a behavioral economic measure of an individual's valuation of future outcomes. While higher DD correlates with obesity in the general adult population, valuation of the future may impact cancer survivors differently due to their unique experiences. We assessed cross-sectional associations between DD, BMI, and healthy lifestyle behaviors in an exploratory analysis of 89 women with hormone receptor positive non-metastatic breast cancer. We found higher DD to be associated with obesity and decreased frequency of vegetable consumption. Future studies should investigate DD as a therapeutic target for novel behavioral interventions in breast cancer survivors affected by obesity. This may improve valuation of the future, increase healthy lifestyle behaviors, and facilitate weight loss to promote overall health and longevity in this population. Obesity in breast cancer (BC) survivors is associated with increased mortality. Delay discounting (DD) is a behavioral economic measure of how individuals value future outcomes. Higher DD correlates with obesity in the general population. Valuation of the future may be associated with obesity differently in cancer survivors. This study evaluated the relationship between DD and obesity in BC survivors. We report an exploratory analysis assessing cross-sectional associations between DD, BMI, and lifestyle behaviors (vegetable and fruit consumption, exercise) related to obesity in 89 women with hormone receptor positive non-metastatic BC. Multivariate linear regression analysis examined demographic and lifestyle behavior variables associated with both BMI and DD. Greater willingness to wait for larger, delayed rewards (lower DD) was significantly associated with lower BMI (standardized beta = -0.32; p < 0.01), independent of age, race, income, time since diagnosis, and menopausal status. There was no significant association between DD and fruit consumption or exercise frequency. Vegetable consumption was significantly associated with lower DD (standardized beta = 0.24; p < 0.05). Higher DD is associated with obesity and decreased frequency of vegetable consumption in BC survivors. Future studies should investigate DD as a therapeutic target for behavioral interventions to facilitate weight loss and promote longevity in this population.
- Privacy-Preserving Synthetic Medical Data Generation with Deep LearningTorfi, Amirsina (Virginia Tech, 2020-08-26)Deep learning models demonstrated good performance in various domains such as ComputerVision and Natural Language Processing. However, the utilization of data-driven methods in healthcare raises privacy concerns, which creates limitations for collaborative research. A remedy to this problem is to generate and employ synthetic data to address privacy concerns. Existing methods for artificial data generation suffer from different limitations, such as being bound to particular use cases. Furthermore, their generalizability to real-world problems is controversial regarding the uncertainties in defining and measuring key realistic characteristics. Hence, there is a need to establish insightful metrics (and to measure the validity of synthetic data), as well as quantitative criteria regarding privacy restrictions. We propose the use of Generative Adversarial Networks to help satisfy requirements for realistic characteristics and acceptable values of privacy metrics, simultaneously. The present study makes several unique contributions to synthetic data generation in the healthcare domain. First, we propose a novel domain-agnostic metric to evaluate the quality of synthetic data. Second, by utilizing 1-D Convolutional Neural Networks, we devise a new approach to capturing the correlation between adjacent diagnosis records. Third, we employ ConvolutionalAutoencoders for creating a robust and compact feature space to handle the mixture of discrete and continuous data. Finally, we devise a privacy-preserving framework that enforcesRényi differential privacy as a new notion of differential privacy.
- Selective regulation of chemosensitivity in glioblastoma by phosphatidylinositol 3-kinase betaPridham, Kevin J.; Hutchings, Kasen R.; Beck, Patrick; Liu, Min; Xu, Eileen; Saechin, Erin; Bui, Vincent; Patel, Chinkal; Solis, Jamie; Huang, Leah; Tegge, Allison; Kelly, Deborah F.; Sheng, Zhi (Elsevier, 2024-06-21)Resistance to chemotherapies such as temozolomide is a major hurdle to effectively treat therapy-resistant glioblastoma. This challenge arises from the activation of phosphatidylinositol 3-kinase (PI3K), which makes it an appealing therapeutic target. However, non-selectively blocking PI3K kinases PI3K⍺/β/𝛿/𝛾 has yielded undesired clinical outcomes. It is, therefore, imperative to investigate individual kinases in glioblastoma’s chemosensitivity. Here,wereport that PI3K kinases were unequally expressed in glioblastoma, with levels of PI3Kβ being the highest. Patients deficient of O6-methylguanine-DNA-methyltransferase(MGMT) and expressing elevated levels of PI3Kβ, defined as MGMT-deficient/PI3Kβ-high, were less responsive to temozolomide and experienced poor prognosis. Consistently, MGMT-deficient/PI3Kβ-high glioblastoma cells were resistant to temozolomide. Perturbation of PI3Kβ, but not other kinases, sensitized MGMTdeficient/ PI3Kβ-high glioblastoma cells or tumors to temozolomide. Moreover, PI3Kβ-selective inhibitors and temozolomide synergistically mitigated the growth of glioblastoma stem cells. Our results have demonstrated an essential role of PI3Kβ in chemoresistance, making PI3Kβ-selective blockade an effective chemosensitizer for glioblastoma.
- Semaglutide and Tirzepatide reduce alcohol consumption in individuals with obesityQuddos, Fatima; Hubshman, Zachary; Tegge, Allison; Sane, Daniel; Marti, Erin; Kablinger, Anita S.; Gatchalian, Kirstin M.; Kelly, Amber L.; DiFeliceantonio, Alexandra G.; Bickel, Warren K. (Nature Portfolio, 2023)Alcohol Use Disorder (AUD) contributes significantly to global mortality. GLP-1 (Glucagon-like peptide-1) and GLP-1/GIP (Glucose-dependent Insulinotropic Polypeptide) agonists, FDA-approved for managing type 2 diabetes and obesity, where the former has shown to effectively reduce the consumption of alcohol in animal models but no reports exist on the latter. In this report, we conducted two studies. In the first study, we conducted an analysis of abundant social media texts. Specifically, a machine-learning based attribution mapping of ~ 68,250 posts related to GLP-1 or GLP-1/GIP agonists on the Reddit platform. Secondly, we recruited participants (n = 153; current alcohol drinkers; BMI ≥ 30) who self-reported either taking Semaglutide (GLP-1 agonist), Tirzepatide (the GLP-1/GIP combination) for ≥ 30 days or, as a control group; no medication to manage diabetes or weight loss for a within and between subject remote study. In the social media study, we report 8 major themes including effects of medications (30%); diabetes (21%); and Weight loss and obesity (19%). Among the alcohol-related posts (n = 1580), 71% were identified as craving reduction, decreased desire to drink, and other negative effects. In the remote study, we observe a significantly lower self-reported intake of alcohol, drinks per drinking episode, binge drinking odds, Alcohol Use Disorders Identification Test (AUDIT) scores, and stimulating, and sedative effects in the Semaglutide or Tirzepatide group when compared to prior to starting medication timepoint (within-subjects) and the control group (between-subjects). In summary, we provide initial real-world evidence of reduced alcohol consumption in people with obesity taking Semaglutide or Tirzepatide medications, suggesting potential efficacy for treatment in AUD comorbid with obesity.
- Understanding family-level decision-making when seeking access to acute surgical care for children: Protocol for a cross-sectional mixed methods studyHall, Bria; Tegge, Allison; Condor, Cesia Cotache; Rhoads, Marie; Wattsman, Terri-Ann; Witcher, Angelica; Creamer, Elizabeth; Tupetz, Anna; Smith, Emily R.; Tokala, Mamata Reddy; Meier, Brian; Rice, Henry E. (PLoS, 2024-06-24)Background There is limited understanding of how social determinants of health (SDOH) impact family decision-making when seeking surgical care for children. Our objectives of this study are to identify key family experiences that contribute to decision-making when accessing surgical care for children, to confirm if family experiences impact delays in care, and to describe differences in family experiences across populations (race, ethnicity, socioeconomic status, rurality). Methods We will use a prospective, cross-sectional, mixed methods design to examine family experiences during access to care for children with appendicitis. Participants will include 242 parents of consecutive children (0–17 years) with acute appendicitis over a 15-month period at two academic health systems in North Carolina and Virginia. We will collect demographic and clinical data. Parents will be administered the Adult Responses to Children’s Symptoms survey (ARCS), the child and parental forms of the Adverse Childhood Experiences (ACE) survey, the Accountable Health Communities Health-Related Social Needs Screening Tool, and Single Item Literacy Screener. Parallel ARCS data will be collected from child participants (8–17 years). We will use nested concurrent, purposive sampling to select a subset of families for semi-structured interviews. Qualitative data will be analyzed using thematic analysis and integrated with quantitative data to identify emerging themes that inform a conceptual model of family-level decision-making during access to surgical care. Multivariate linear regression will be used to determine association between the appendicitis perforation rate and ARCS responses (primary outcome). Secondary outcomes include comparison of health literacy, ACEs, and SDOH, clinical outcomes, and family experiences across populations. Discussion We expect to identify key family experiences when accessing care for appendicitis which may impact outcomes and differ across populations. Increased understanding of how SDOH and family experiences influence family decision-making may inform novel strategies to mitigate surgical disparities in children.
- Variable selection for generalized linear mixed models and non-Gaussian Genome-wide associated study dataXu, Shuangshuang (Virginia Tech, 2024-06-11)Genome-wide associated study (GWAS) aims to identify associated single nucleotide polymorphisms (SNP) for phenotypes. SNP has the characteristic that the number of SNPs is from hundred of thousands to millions. If p is the number of SNPs and n is the sample size, it is a p>>n variable selection problem. To solve this p>>n problem, the common method for GWAS is single marker analysis (SMA). However, since SNPs are highly correlated, SMA identifies true causal SNPs with high false discovery rate. In addition, SMA does not consider interaction between SNPs. In this dissertation, we propose novel Bayesian variable selection methods BG2 and IBG3 for non-Gaussian GWAS data. To solve ultra-high dimension problem and highly correlated SNPs problem, BG2 and IBG3 have two steps: screening step and fine-mapping step. In the screening step, BG2 and IBG3, like SMA method, only have one SNP in one model and screen to obtain a subset of most associated SNPs. In the fine-mapping step, BG2 and IBG3 consider all possible combinations of screened candidate SNPs to find the best model. Fine-mapping step helps to reduce false positives. In addition, IBG3 iterates these two steps to detect more SNPs with small effect size. In simulation studies, we compare our methods with SMA methods and fine-mapping methods. We also compare our methods with different priors for variables, including nonlocal prior, unit information prior, Zellner-g prior, and Zellner-Siow prior. Our methods are applied to substance use disorder (alcohol comsumption and cocaine dependence), human health (breast cancer), and plant science (the number of root-like structure).