publications

2025

1. Interpretable Model Summaries Using the Wasserstein Distance

   Eric Arthur Dunipace, and Lorenzo Trippa

   The American Statistician, 2025

   DOI

2022

1. Optimal transport weights for causal inference

   Eric Dunipace

   arXiv:2109.01991 [cs, econ, stat], 2022

   arXiv: 2109.01991

   Abs

   Weighting methods are a common tool to de-bias estimates of causal effects. And though there are an increasing number of seemingly disparate methods, many of them can be folded into one unifying regime: causal optimal transport. This new method directly targets distributional balance by minimizing optimal transport distances between treatment and control groups or, more generally, between a source and target population. Our approach is model-free but can also incorporate moments or any other important functions of covariates that the researcher desires to balance. We find that the causal optimal transport outperforms competitor methods when both the propensity score and outcome models are misspecified, indicating it is a robust alternative to common weighting methods. Finally, we demonstrate the utility of our method in an external control study examining the effect of misoprostol versus oxytocin for treatment of post-partum hemorrhage.

2. Association Between Periodontal Disease and Cardiovascular Disease Using Publicly Available Data from the National Health and Nutrition Examination Survey (NHANES)

   Kyari Sumayin Ngamdu, Indika Mallawaarachchi, Eric A. Dunipace, and 5 more authors

   The American Journal of Cardiology, 2022

2021

1. {approxOT}: approximate optimal transport

   Eric A. Dunipace

   2021

2019

1. Interpretable Posterior Summaries Using the Wasserstein Distance

   Eric Dunipace, and Lorenzo Trippa

   2019

   Place: Denver, CO, USA

   Abs

   In the current computing age, models can often have hundreds or even thousands of parameters. With these large models comes the risk of losing the ability to communicate and understand the precise meaning of the individual parameters in a model. In a frequentist setting, one can use an L1 penalty to reduce the number of parameters in a model but similar methods have not been developed for Bayesian settings where the quantity of interest involves an integral over the posterior distribution. We introduce a new method using a penalized 2-Wasserstein distance to reduce the dimensionality of the parameter space while still obtaining a distribution over the remaining dimensions. Our method allows users to select a budget on how many parameters they wish to understand, interpret, and communicate to an audience, and, in a data dependent way, select a reduced posterior of the selected dimension that minimizes the distance to the full posterior. We provide simulation results comparing performance to other posterior summary methods and apply the method to cancer and environmental health data.

2018

1. Liver Disease as a Predictor of New Onset Atrial Fibrillation

   William A Huang, Eric A Dunipace, Julie M Sorg, and 1 more author

   Journal of the American Heart Association, 2018

2014

1. TraumaLink: Helping People Survive Traffic Accidents in Bangladesh

   Jon Moussally, Eric Dunipace, and Ryan Fu

   In {Health and South Asia}, 2014

2011

1. Differential requirements for NAIP5 in activation of the NLRC4 inflammasome

   K.L. Lightfield, J. Persson, N.J. Trinidad, and 7 more authors

   Infection and Immunity, 2011

   Abs DOI

   Inflammasomes are cytosolic multiprotein complexes that assemble in response to infectious or noxious stimuli and activate the CASPASE-1 protease. The inflammasome containing the nucleotide binding domainleucine-rich repeat (NBD-LRR) protein NLRC4 (interleukin-converting enzyme protease-activating factor [IPAF]) responds to the cytosolic presence of bacterial proteins such as flagellin or the inner rod component of bacterial type III secretion systems (e.g., Salmonella PrgJ). In some instances, such as infection with Legionella pneumophila, the activation of the NLRC4 inflammasome requires the presence of a second NBDLRR protein, NAIP5. NAIP5 also is required for NLRC4 activation by the minimal C-terminal flagellin peptide, which is sufficient to activate NLRC4. However, NLRC4 activation is not always dependent upon NAIP5. In this report, we define the molecular requirements for NAIP5 in the activation of the NLRC4 inflammasome. We demonstrate that the N terminus of flagellin can relieve the requirement for NAIP5 during the activation of the NLRC4 inflammasome. We also demonstrate that NLRC4 responds to the Salmonella protein PrgJ independently of NAIP5. Our results indicate that NAIP5 regulates the apparent specificity of the NLRC4 inflammasome for distinct bacterial ligands. © 2011, American Society for Microbiology.

2008

1. Critical function for Naip5 in inflammasome activation by a conserved carboxy-terminal domain of flagellin

   K.L. Lightfield, J. Persson, S.W. Brubaker, and 10 more authors

   Nature Immunology, 2008

   Abs DOI

   Inflammasomes are cytosolic multiprotein complexes that sense microbial infection and trigger cytokine production and cell death. However, the molecular components of inflammasomes and what they sense remain poorly defined. Here we demonstrate that 35 amino acids of the carboxyl terminus of flagellin triggered inflammasome activation in the absence of bacterial contaminants or secretion systems. To further elucidate the host flagellin-sensing pathway, we generated mice deficient in the intracellular sensor Naip5. These mice failed to activate the inflammasome in response to the 35 amino acids of flagellin or in response to Legionella pneumophila infection. Our data clarify the molecular basis for the cytosolic response to flagellin.