A5055014782- Gene expression and cancer classification
- Bioinformatics and Genomic Networks
- Statistical Methods and Inference
- Bayesian Methods and Mixture Models
- Iron Metabolism and Disorders
- Statistical Methods and Bayesian Inference
- Tensor decomposition and applications
- Cancer Genomics and Diagnostics
- Pneumocystis jirovecii pneumonia detection and treatment
- Hemoglobinopathies and Related Disorders
- Metabolomics and Mass Spectrometry Studies
- Peptidase Inhibition and Analysis
- Primary Care and Health Outcomes
- Infant Nutrition and Health
- Genetic Associations and Epidemiology
- Epigenetics and DNA Methylation
- Genetic factors in colorectal cancer
- Chronic Disease Management Strategies
- Pharmacological Effects and Toxicity Studies
- Chronic Obstructive Pulmonary Disease (COPD) Research
- Mitochondrial Function and Pathology
- Gut microbiota and health
- Breastfeeding Practices and Influences
- Computational Drug Discovery Methods
- Genetic and phenotypic traits in livestock
University of Minnesota
2016-2025
University of Minnesota System
2019-2024
Twin Cities Orthopedics
2024
Minnesota Department of Health
2024
Minneapolis Institute of Arts
2024
Duke Medical Center
2013-2015
Center for Human Genetics
2013-2014
Duke University
2013-2014
University of North Carolina at Chapel Hill
2010-2013
Duke University Hospital
2013
Research in several fields now requires the analysis of data sets which multiple high-dimensional types are available for a common set objects. In particular, The Cancer Genome Atlas (TCGA) includes from diverse genomic technologies on same cancerous tumor samples. this paper we introduce Joint and Individual Variation Explained (JIVE), general decomposition variation integrated such sets. consists three terms: low-rank approximation capturing joint across types, approximations structured...
The task of clustering a set objects based on multiple sources data arises in several modern applications. We propose an integrative statistical model that permits separate the for each source. These clusterings adhere loosely to overall consensus clustering, and hence they are not independent. describe computationally scalable Bayesian framework simultaneous estimation both source-specific clusterings. demonstrate this flexible approach is more robust than joint all sources, powerful source...
Abstract While gut microbiome and host gene regulation independently contribute to gastrointestinal disorders, it is unclear how the two may interact influence pathophysiology. Here we developed a machine learning-based framework jointly analyse paired transcriptomic ( n = 208) profiles from colonic mucosal samples of patients with colorectal cancer, inflammatory bowel disease irritable syndrome. We identified associations between microbes genes that depict shared as well disease-specific...
We propose a framework for the linear prediction of multi-way array (i.e., tensor) from another arbitrary dimension, using contracted tensor product. This generalizes several existing approaches, including methods to predict scalar outcome tensor, matrix matrix, or scalar. describe an approach that exploits multiway structure both predictors and outcomes by restricting coefficients have reduced CP-rank. general efficient algorithm penalized least-squares estimation, which allows ridge (L2)...
Scientists and regulators are often faced with complex decisions, where use of scarce resources must be prioritized using collections diverse information. The Toxicological Prioritization Index (ToxPi™) was developed to enable integration multiple sources evidence on exposure and/or safety, transformed into transparent visual rankings facilitate decision making. associated graphical profiles can used prioritize in various contexts, such as testing chemical toxicity or assessing similarity...
: The integrative analysis of multiple high-throughput data sources that are available for a common sample set is an increasingly goal in biomedical research. Joint and individual variation explained (JIVE) tool exploratory dimension reduction decomposes multi-source dataset into three terms: low-rank approximation capturing joint across sources, approximations structured to each source residual noise. JIVE has been used explore variety application areas but its accessibility was previously...
Trichloroethylene (TCE) is a widely used industrial chemical and common environmental contaminant. It well-known carcinogen in rodents probable humans. Studies utilizing panels of mouse inbred strains afford unique opportunity to understand both metabolic genetic basis for differences responses TCE. We tested the hypothesis that strain- liver-specific toxic effects TCE are genetically controlled mechanisms toxicity susceptibility can be uncovered by exploring using diverse panel strains....
A shift in toxicity testing from vivo to vitro may efficiently prioritize compounds, reveal new mechanisms, and enable predictive modeling. Quantitative high-throughput screening (qHTS) is a major source of data for computational toxicology, our goal this study was aid the development models chemical-induced toxicity, anchored on interindividual genetic variability. Eighty-one human lymphoblast cell lines 27 Centre d'Etude du Polymorphisme Humain trios were exposed 240 chemical substances...
Iron deficiency (ID) anemia leads to long-term neurodevelopmental deficits by altering iron-dependent brain metabolism. The objective of the study was determine if ID induces metabolomic abnormalities in cerebrospinal fluid (CSF) pre-anemic stage and ascertain aspects abnormal metabolism affected.Standard hematological parameters [hemoglobin (Hgb), mean corpuscular volume (MCV), transferrin (Tf) saturation, zinc protoporphyrin/heme (ZnPP/H)] were compared at 2, 4, 6, 8, 12 months...
In Idiopathic Pulmonary Fibrosis (IPF), there is unrelenting scarring of the lung mediated by pathological mesenchymal progenitor cells (MPCs) that manifest autonomous fibrogenicity in xenograft models. To determine where along their differentiation trajectory IPF MPCs acquire fibrogenic properties, we analyzed transcriptome 335 isolated from lungs 3 control and patients at single-cell level. Using transcriptional entropy as a metric for differentiated state, found least displayed largest...
Human milk is a complex mix of nutritional and bioactive components that provide complete nutrition for the infant. However, we lack systematic knowledge factors shaping composition how variation influences infant health. Here, used multi-omic profiling to characterize interactions between maternal genetics, gene expression, composition, fecal microbiome in 242 exclusively breastfeeding mother-infant pairs. We identified 487 genetic loci associated with expression unique lactating mammary...
Objective We aimed to quantify differences in the brain and spinal cord between Friedreich ataxia controls, stratified by age disease stage, including for first time young children. Methods TRACK‐FA is largest prospective, longitudinal, multi‐modal neuroimaging study date. assessed individuals with 5 42 years, at 7 sites across 4 continents. The 17 imaging primary outcome measures (POMs) were selected from metrics that showed a significant longitudinal change previous small‐scale studies....
Several modern applications require the integration of multiple large data matrices that have shared rows and/or columns. For example, cancer studies integrate omics platforms across types cancer, pan-omics pan-cancer analysis, extended our knowledge molecular heterogeneity beyond what was observed in single tumor and platform studies. However, these been limited by available statistical methodology. We propose a flexible approach to simultaneous factorization decomposition variation such...
Importance Gestational diabetes (GD) is linked to health risks for the birthing parent and infant. The outcomes of GD on human milk composition are mostly unknown. Objective To determine associations between GD, metabolome, infant growth body composition. Design, Setting, Participants Cohort study using data from Mothers Infants Linked Healthy Growth Maternal Milk, Metabolism, Microbiome studies at University Oklahoma Minnesota, large prospective US cohorts with a high proportion exclusive...
Systems vaccinology studies have been used to build computational models that predict individual vaccine responses and identify the factors contributing differences in outcome. Comparing such is challenging due variability study designs. To address this, we established a community resource compare predicting B. pertussis booster generate experimental data for explicit purpose of model evaluation. We here describe our second prediction challenge using this resource, where benchmarked 49...
Exercise is recommended for postpartum health, but its impacts on breastmilk composition and offspring are understudied. To test whether the metabolome altered with (i) acute exercise and/or (ii) habitual physical activity, (iii) exercise-altered metabolites associated infant adiposity. Milk were assessed before after in association activity score two independent cohorts. Two academic medical centers. The cohort had 15 mother-infant dyads. nested case-control analysis 84 physically active...
Chiari Type I Malformation (CMI) is characterized by herniation of the cerebellar tonsils through foramen magnum at base skull, resulting in significant neurologic morbidity. As CMI patients display a high degree clinical variability and multiple mechanisms have been proposed for tonsillar herniation, it hypothesized that this heterogeneous disorder due to genetic environmental factors. The purpose present study was gain better understanding what factors contribute heterogeneity using an...
Joint and Individual Variation Explained (JIVE) is used for the integrated unsupervised analysis of metabolomic profiles from multiple data sources.
Chronic obstructive pulmonary disease (COPD) is a known risk factor for developing lung cancer but the underlying mechanisms remain unknown. We hypothesise that COPD stroma contains molecular supporting tumourigenesis. conducted an unbiased multi-omic analysis to identify gene expression patterns distinguish in patients with or without cancer. obtained tissue from and (tumour adjacent non-malignant tissue) those profiling of proteomic mRNA (both cytoplasmic polyribosomal). used Joint...
Predictive modeling from high-dimensional genomic data is often preceded by a dimension reduction step, such as principal component analysis (PCA). However, the application of PCA not straightforward for multisource data, wherein multiple sources ‘omics measure different but related biological components. In this article, we use recent advances in predictive modeling. particular, apply exploratory results Joint and Individual Variation Explained (JIVE), an extension prediction differing...
OBJECTIVE Identify the improvement in diabetes performance measures and population-based clinical outcomes resulting from changes care management processes (CMP) primary practices over 3 years. RESEARCH DESIGN AND METHODS This repeated cross-sectional study tracked for all patients seen a cohort of 330 2017 2019. Unit analysis was patient-year with practice-level CMP exposures. Causal inference is based on dynamic individual CMPs between years by practice. We used Bayesian method to...