Bayesian Variable Selection and Model Choice for Generalized Additive Mixed Models

Bayesian variable selection, model choice, and regularized estimation for (spatial) generalized additive mixed regression models via stochastic search variable selection with spike-and-slab priors.

Parallel Linear Mixed Model

Embarrassingly Parallel Linear Mixed Model calculations spread across local cores which repeat until convergence.

Some Algorithms for Mixed Models

This program can be used to fit Gaussian linear mixed models (LMM). Univariate and multivariate response models, multiple variance components, as well as, certain correlation and covariance structures are supported. In many occasions, the user can pick one of the several mixed model fitting algorithms, which are explained further in the details section. Some algorithms are specific to certain types of models (univariate or multivariate, diagonal or non-diagonal residual, one or multiple variance components, etc,...).

Bayesian Mixing Models in R

Creates and runs Bayesian mixing models to analyze biological tracer data (i.e. stable isotopes, fatty acids), which estimate the proportions of source (prey) contributions to a mixture (consumer). 'MixSIAR' is not one model, but a framework that allows a user to create a mixing model based on their data structure and research questions, via options for fixed/ random effects, source data types, priors, and error terms. 'MixSIAR' incorporates several years of advances since 'MixSIR' and 'SIAR'.

Kernel Ridge Mixed Model

Solves kernel ridge regression, within the the mixed model framework, for the linear, polynomial, Gaussian, Laplacian and ANOVA kernels. The model components (i.e. fixed and random effects) and variance parameters are estimated using the expectation-maximization (EM) algorithm. All the estimated components and parameters, e.g. BLUP of dual variables and BLUP of random predictor effects for the linear kernel (also known as RR-BLUP), are available. The kernel ridge mixed model (KRMM) is described in Jacquin L, Cao T-V and Ahmadi N (2016) A Unified and Comprehensible View of Parametric and Kernel Methods for Genomic Prediction with Application to Rice. Front. Genet. 7:145. .

Tidying Methods for Mixed Models

Convert fitted objects from various R mixed-model packages into tidy data frames along the lines of the 'broom' package. The package provides three S3 generics for each model: tidy(), which summarizes a model's statistical findings such as coefficients of a regression; augment(), which adds columns to the original data such as predictions, residuals and cluster assignments; and glance(), which provides a one-row summary of model-level statistics.

A Stable Isotope Mixing Model

Fits Stable Isotope Mixing Models (SIMMs) and is meant as a longer term replacement to the previous widely-used package SIAR. SIMMs are used to infer dietary proportions of organisms consuming various food sources from observations on the stable isotope values taken from the organisms' tissue samples. However SIMMs can also be used in other scenarios, such as in sediment mixing or the composition of fatty acids. The main functions are simmr_load() and simmr_mcmc(). The two vignettes contain a quick start and a full listing of all the features. The methods used are detailed in the papers Parnell et al 2010 , and Parnell et al 2013 .

Stable Isotope Mixing Model

Estimates diet contributions from isotopic sources using JAGS. Includes estimation of concentration dependence and measurement error.

Latent Factor Mixed Models

Fast and accurate inference of gene-environment associations (GEA) in genome-wide studies (Caye et al., 2019, ). We developed a least-squares estimation approach for confounder and effect sizes estimation that provides a unique framework for several categories of genomic data, not restricted to genotypes. The speed of the new algorithm is several times faster than the existing GEA approaches, then our previous version of the 'LFMM' program present in the 'LEA' package (Frichot and Francois, 2015, ).

GEMMA Multivariate Linear Mixed Model

Fits a multivariate linear mixed effects model that uses a polygenic term, after Zhou & Stephens (2014) (< https://www.nature.com/articles/nmeth.2848>). Of particular interest is the estimation of variance components with restricted maximum likelihood (REML) methods. Genome-wide efficient mixed-model association (GEMMA), as implemented in the package 'gemma2', uses an expectation-maximization algorithm for variance components inference for use in quantitative trait locus studies.