Gaussian process modelling in approximate Bayesian computation to estimate horizontal gene transfer in bacteria

DOI10.1214/18-AOAS1150zbMATH Open1411.62320arXiv1610.06462WikidataQ128949382 ScholiaQ128949382MaRDI QIDQ1728645FDOQ1728645

Authors: Marko Järvenpää, Michael U. Gutmann, Pekka Marttinen, Aki Vehtari 

Publication date: 25 February 2019

Published in: The Annals of Applied Statistics (Search for Journal in Brave)

Abstract: Approximate Bayesian computation (ABC) can be used for model fitting when the likelihood function is intractable but simulating from the model is feasible. However, even a single evaluation of a complex model may take several hours, limiting the number of model evaluations available. Modelling the discrepancy between the simulated and observed data using a Gaussian process (GP) can be used to reduce the number of model evaluations required by ABC, but the sensitivity of this approach to a specific GP formulation has not yet been thoroughly investigated. We begin with a comprehensive empirical evaluation of using GPs in ABC, including various transformations of the discrepancies and two novel GP formulations. Our results indicate the choice of GP may significantly affect the accuracy of the estimated posterior distribution. Selection of an appropriate GP model is thus important. We formulate expected utility to measure the accuracy of classifying discrepancies below or above the ABC threshold, and show that it can be used to automate the GP model selection step. Finally, based on the understanding gained with toy examples, we fit a population genetic model for bacteria, providing insight into horizontal gene transfer events within the population and from external origins.

Full work available at URL: https://arxiv.org/abs/1610.06462

Recommendations

• Efficient acquisition rules for model-based approximate Bayesian computation
• Approximate Bayesian computation with differential evolution
• Hierarchical approximate Bayesian computation
• Extending approximate Bayesian computation methods to high dimensions via a Gaussian copula model
• Improved approximate Bayesian computation methods via empirical likelihood

zbMATH Keywords

model selectionapproximate Bayesian computationGaussian processintractable likelihoodinput-dependent noise

Mathematics Subject Classification ID

Bayesian inference (62F15) Applications of statistics to biology and medical sciences; meta analysis (62P10) Gaussian processes (60G15) Genetics and epigenetics (92D10)

Cites Work

• Bayesian Synthetic Likelihood
• A survey of Bayesian predictive methods for model assessment, selection and comparison
• Bayesian Model Assessment and Comparison Using Cross-Validation Predictive Densities
• Gaussian processes for machine learning.
• Nonparametric guidance of autoencoder representations using label information
• Bayesian indirect inference using a parametric auxiliary model
• Bayesian optimization for likelihood-free inference of simulator-based statistical models
• Constructing Summary Statistics for Approximate Bayesian Computation: Semi-Automatic Approximate Bayesian Computation
• Approximate Bayesian computation: a nonparametric perspective
• Sequential Monte Carlo without likelihoods
• GPstuff: Bayesian modeling with Gaussian processes
• Approximate Bayesian computational methods
• Adaptive approximate Bayesian computation
• Estimation of parameters for macroparasite population evolution using approximate Bayesian computation
• An adaptive sequential Monte Carlo method for approximate Bayesian computation
• Adaptive approximate Bayesian computation for complex models
• A tutorial on approximate Bayesian computation
• Accelerating pseudo-marginal MCMC using Gaussian processes
• Gaussian process modelling in approximate Bayesian computation to estimate horizontal gene transfer in bacteria
• Efficient acquisition rules for model-based approximate Bayesian computation

Cited In (9)

• Improved approximate Bayesian computation methods via empirical likelihood
• Scalable Bayesian inference for the inverse temperature of a hidden Potts model
• Gaussian process modelling in approximate Bayesian computation to estimate horizontal gene transfer in bacteria
• Efficient acquisition rules for model-based approximate Bayesian computation
• Parallel Gaussian process surrogate Bayesian inference with noisy likelihood evaluations
• New approximate Bayesian computation algorithm for censored data
• Adaptive approximate Bayesian computation tolerance selection
• Bayesian optimization of functional output in inverse problems
• Uncertainty quantification in deterministic parameterization of single diode model of a silicon solar cell

Uses Software

• GPstuff
• GPS-ABC
• Spearmint

This page was built for publication: Gaussian process modelling in approximate Bayesian computation to estimate horizontal gene transfer in bacteria

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1728645)