A Novel Algorithmic Approach to Bayesian Logic Regression (with Discussion)

@article{Hubin2020ANA,
  title={A Novel Algorithmic Approach to Bayesian Logic Regression (with Discussion)},
  author={Aliaksandr Hubin and Geir Storvik and Florian Frommlet},
  journal={Bayesian Analysis},
  year={2020}
}
Logic regression was developed more than a decade ago as a tool to construct predictors from Boolean combinations of binary covariates. It has been mainly used to model epistatic effects in genetic association studies, which is very appealing due to the intuitive interpretation of logic expressions to describe the interaction between genetic variations. Nevertheless logic regression has remained less well known than other approaches to epistatic association mapping. Here we will adopt an… 

Tables from this paper

Deep Bayesian regression models
TLDR
The ideas of deep learning are brought into a statistical framework which yields more parsimonious models and allows to quantify model uncertainty, and the class of deep Bayesian regression models (DBRM) is introduced consisting of a generalized linear model combined with a comprehensive non-linear feature space.
Flexible Bayesian Nonlinear Model Configuration
TLDR
This paper introduces a flexible approach for the construction and selection of highly  flexible nonlinear parametric regression models, and illustrates how this approach is used to obtain meaningful nonlinear models.
A subsampling approach for Bayesian model selection
TLDR
A version of a popular batch stochastic gradient descent algorithm is suggested for estimating the marginal likelihood of a GLM by subsampling from the data and is combined with Markov chain Monte Carlo based methods for Bayesian model selection and provides some theoretical results on convergence of the estimates.
A Bayesian Binomial Regression Model with Latent Gaussian Processes for Modelling DNA Methylation
TLDR
This work incorporates dependence on the covariates and the spatial dependence of the methylation probability for observations from a pool of cells by means of a binomial regression model with a latent Gaussian field and a logit link function and applies a Bayesian approach including prior specifications on model configurations.
Combining Model and Parameter Uncertainty in Bayesian Neural Networks
TLDR
This paper introduces the concept of model uncertainty in BNNs and suggests an adaptation of a scalable variational inference approach with reparametrization of marginal inclusion probabilities to incorporate the model space constraints and shows that incorporating model uncertainty via Bayesian model averaging andBayesian model selection allows to drastically sparsify the structure of BNN's.
Configurational Causal Modeling and Logic Regression.
TLDR
It turns out that CCMs and LRMs have complementary strengths and weaknesses, which creates various promising avenues for cross-validation.
An adaptive simulated annealing EM algorithm for inference on non-homogeneous hidden Markov models
  • A. Hubin
  • Mathematics, Computer Science
    AIIPCC '19
  • 2019
TLDR
An adaptive (A) simulated annealing (SA) expectation maximization (EM) algorithm (ASA-EM) is proposed for joint optimization of models and their parameters with respect to a criterion of interest.
Using logic regression to characterize extreme heat exposures and their health associations: a time-series study of emergency department visits in Atlanta
TLDR
This work describes how logic regression can be utilized as a data-driven approach to identify extreme heat exposure definitions using health outcome data to help improve future heat warning systems and response plans.
Reversible Genetically Modified Mode Jumping MCMC
TLDR
A reversible version of a genetically modified mode jumping Markov chain Monte Carlo algorithm for inference on posterior model probabilities in complex model spaces, where the number of explanatory variables is prohibitively large for classical Markov Chain Monte Carlo methods.
M E ] 1 1 O ct 2 02 1 Reversible Genetically Modified Mode Jumping MCMC
  • 2021

References

SHOWING 1-10 OF 64 REFERENCES
Detection of epistatic effects with logic regression and a classical linear regression model
TLDR
A logic regression framework is proposed to detect higher order logic interactions in logic regression models leads to a significant increase of power to detect such interactions as compared to a Cockerham’s approach.
Comparing Logic Regression Based Methods for Identifying SNP Interactions
TLDR
This paper compares the performance in identifying SNP interactions associated with the case-control status of the three logic regression based methods and stepwise logistic regression in a simulation study and in a study of breast cancer.
Logic Regression
Logic regression is an adaptive regression methodology that attempts to construct predictors as Boolean combinations of binary covariates. In many regression problems a model is developed that
Logic regression and its extensions.
Identifying interacting SNPs using Monte Carlo logic regression
TLDR
Monte Carlo logic regression is proposed, an exploratory tool that combines Markov chain Monte Carlo and logic regression, an adaptive regression methodology that attempts to construct predictors as Boolean combinations of binary covariates such as SNPs.
QTL Mapping Using a Memetic Algorithm with Modifications of BIC as Fitness Function
TLDR
A memetic algorithm (MA) is introduced to find the model which minimizes the selection criterion and it turns out that MA in combination with mBIC2 performs best, where determining QTL positions based on marker posterior probabilities yields even better results than using the best model selected by MA.
Deep Bayesian regression models
TLDR
The ideas of deep learning are brought into a statistical framework which yields more parsimonious models and allows to quantify model uncertainty, and the class of deep Bayesian regression models (DBRM) is introduced consisting of a generalized linear model combined with a comprehensive non-linear feature space.
Identification of SNP interactions using logic regression.
TLDR
This paper shows how logic regression can be employed to identify SNP interactions explanatory for the disease status in a case-control study and proposes 2 measures for quantifying the importance of these interactions for classification.
A Bayesian Binomial Regression Model with Latent Gaussian Processes for Modelling DNA Methylation
TLDR
This work incorporates dependence on the covariates and the spatial dependence of the methylation probability for observations from a pool of cells by means of a binomial regression model with a latent Gaussian field and a logit link function and applies a Bayesian approach including prior specifications on model configurations.
Extended Bayesian information criteria for model selection with large model spaces
TLDR
This paper re-examine the Bayesian paradigm for model selection and proposes an extended family of Bayesian information criteria, which take into account both the number of unknown parameters and the complexity of the model space.
...
...