Parallelize Common Functions via One Magic Function

The futurize() function turns sequential map-reduce functions such as base::lapply(), purrr::map(), 'foreach::foreach() %do% { ... }' into concurrent alternatives, providing you with a simple, straightforward path to scalable parallel computing via the 'future' ecosystem . By combining this transpiler function with R's native pipe operator, you have a convenient way for speeding up iterative computations with minimal refactoring, e.g. 'lapply(xs, fcn) |> futurize()', 'purrr::map(xs, fcn) |> futurize()', and 'foreach::foreach(x = xs) %do% { fcn(x) } |> futurize()'. Other map-reduce packages that can be "futurized" are 'BiocParallel', 'plyr', 'crossmap', 'pbapply' packages. There is also support for a growing set of domain-specific packages on CRAN (e.g. 'boot', 'caret', 'DiceKriging', 'ez', 'fgsea', 'fwb', 'gamlss', 'glmmTMB', 'glmnet', 'kernelshap', 'lme4', 'metafor', 'mgcv', 'modelsummary', 'parameters', 'partykit', 'pls', 'pvclust', 'riskRegression', 'rugarch', 'sandwich', 'seriation', 'shapr', 'Sim.DiffProc', 'SimDesign', 'stars', 'strucchange', 'SuperLearner', 'terra', 'tm', 'TSP', and 'vegan') and on Bioconductor (e.g. 'DESeq2', 'GenomicAlignments', 'GSVA', 'Rsamtools', 'scater', 'scuttle', 'SingleCellExperiment', and 'sva').

Additive Mixed Meta-Analysis with Spline Meta-Regression

Fit additive mixed meta-analysis (AMMA) models, extending the 'mixmeta' package < https://cran.r-project.org/package=mixmeta> to allow for spline-based meta-regression. Functions combine features of 'mgcv' < https://cran.r-project.org/package=mgcv> for building spline components and 'mixmeta' for estimating general mixed-effects meta-analysis models.

Visualisations for Generalized Additive Models

Extension of the 'mgcv' package, providing visual tools for Generalized Additive Models that exploit the additive structure of such models, scale to large data sets and can be used in conjunction with a wide range of response distributions. The focus is providing visual methods for better understanding the model output and for aiding model checking and development beyond simple exponential family regression. The graphical framework is based on the layering system provided by 'ggplot2'.

Discrete Prolate Spheroidal (Slepian) Sequence Regression Smoothers

Interface for creation of 'slp' class smoother objects for use in Generalized Additive Models (as implemented by packages 'gam' and 'mgcv').

Censored Regression with Smooth Terms

Implementation of Tobit type I and type II families for censored regression using the 'mgcv' package, based on methods detailed in Woods (2016) .

Tidy Prediction and Plotting of Generalised Additive Models

Provides functions that compute predictions from Generalised Additive Models (GAMs) fitted with 'mgcv' and return them as a tibble. These can be plotted with a generic plot()-method that uses 'ggplot2' or plotted as any other data frame. The main function is predict_gam().

Fast Robust Additive Models using Gamma Divergence

Provides a computationally efficient method (at least at the time of development!) for robust additive modeling based on the gamma divergence, implementing the methodology of Hui, Ding, and Sugasawa (2026) . The method assumes an identity link and normally distributed errors, and applies the gamma divergence to the resulting restricted maximum likelihood function so as to obtain a loss function that is less sensitive to outlying responses. Because 'gdam' uses a model fitted via the 'mgcv' package as the starting point for optimization, and leverages many of its existing techniques for estimation and inference, users can take advantage of many of the smoothing options available in the 'mgcv' package for constructing additive models.

Hierarchical Partitioning of Adjusted R2 and Explained Deviance for Generalized Additive Models

Conducts hierarchical partitioning to calculate individual contributions of each predictor towards adjusted R2 and explained deviance for generalized additive models based on output of 'gam()' and 'bam()' in 'mgcv' package, applying the algorithm in this paper: Lai(2024) .

Item-by-Item Iterative Model Fitting

Models high-dimensional data, such as RNA-seq or proteomic data using an item-by-item strategy. The package contains functions to wrap high-dimensional data and iterate over them using established R packages for regression modelling (e.g., 'glmmTMB' or 'mgcv').

Penalized Splines Mixed-Effects Models

Fit penalized splines mixed-effects models (a special case of additive models) for large longitudinal datasets. The package includes a psme() function that (1) relies on package 'mgcv' for constructing population and subject smooth functions as penalized splines, (2) transforms the constructed additive model to a linear mixed-effects model, (3) exploits package 'lme4' for model estimation and (4) backtransforms the estimated linear mixed-effects model to the additive model for interpretation and visualization. See Pedersen et al. (2019) and Bates et al. (2015) for an introduction. Unlike the gamm() function in 'mgcv', the psme() function is fast and memory-efficient, able to handle datasets with millions of observations.