Dataset Information

The Hamming Ball Sampler.

ABSTRACT: We introduce the Hamming ball sampler, a novel Markov chain Monte Carlo algorithm, for efficient inference in statistical models involving high-dimensional discrete state spaces. The sampling scheme uses an auxiliary variable construction that adaptively truncates the model space allowing iterative exploration of the full model space. The approach generalizes conventional Gibbs sampling schemes for discrete spaces and provides an intuitive means for user-controlled balance between statistical efficiency and computational tractability. We illustrate the generic utility of our sampling algorithm through application to a range of statistical models. Supplementary materials for this article are available online.

SUBMITTER: Titsias MK

PROVIDER: S-EPMC5796496 | biostudies-literature | 2017

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

The Hamming Ball Sampler.

Titsias Michalis K MK Yau Christopher C

Journal of the American Statistical Association 20170903 520

We introduce the Hamming ball sampler, a novel Markov chain Monte Carlo algorithm, for efficient inference in statistical models involving high-dimensional discrete state spaces. The sampling scheme uses an auxiliary variable construction that adaptively truncates the model space allowing iterative exploration of the full model space. The approach generalizes conventional Gibbs sampling schemes for discrete spaces and provides an intuitive means for user-controlled balance between statistical ef ...[more]

PMID: 29456460

Similar Datasets

Project description:There is growing interest in studying the toxicity and health risk of exposure to multi-pollutant mixtures found in ambient air, and the U.S. Environmental Protection Agency (EPA) is moving towards setting standards for these types of mixtures. Additionally, the Health Effects Institute's strategic plan aims to develop and apply next-generation multi-pollutant approaches to understanding the health effects of air pollutants. There's increasing concern that conventional in vitro exposure methods are not adequate to meet EPA's strategic plan to demonstrate a direct link between air pollution and health effects. To meet the demand for new in vitro technology that better represents direct air-to-cell inhalation exposures, a new system that exposes cells at the air-liquid interface was developed. This new system, named the Gillings Sampler, is a modified two-stage electrostatic precipitator that provides a viable environment for cultured cells. Polystyrene latex spheres were used to determine deposition efficiencies (38-45%), while microscopy and imaging techniques were used to confirm uniform particle deposition. Negative control A549 cell exposures indicated the sampler can be operated for up to 4h without inducing any significant toxic effects on cells, as measured by lactate dehydrogenase (LDH) and interleukin-8 (IL-8). A novel positive aerosol control exposure method, consisting of a p-tolualdehyde (TOLALD) impregnated mineral oil aerosol (MOA), was developed to test this system. Exposures to the toxic MOA at a 1 ng/cm(2) dose of TOLALD yielded a reproducible 1.4 and 2-fold increase in LDH and IL-8 mRNA levels over controls. This new system is intended to be used as an alternative research tool for aerosol in vitro exposure studies. While further testing and optimization is still required to produce a "commercially ready" system, it serves as a stepping-stone in the development of cost-effective in vitro technology that can be made accessible to researchers in the near future.

Dataset Information

The Hamming Ball Sampler.

Publications

The Hamming Ball Sampler.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure