Causal Inference

causaldef v0.2.0: Implements Le Cam deficiency theory for causal inference, as described in Akdemir (2026). Provides theorem-backed bounds together with computable proxy diagnostics for information loss from confounding, selection bias, and distributional shift. Supports continuous, binary, count, survival, and competing risks outcomes. Key features include propensity-score total-variation deficiency proxies, negative control diagnostics, policy regret bounds, and sensitivity analysis via confounding frontiers. There are eleven vignettes including Introduction and Survival Analysis.

CausalSpline v0.1.0: Estimates nonlinear causal dose-response functions for continuous treatments using spline-based methods under standard causal assumptions (unconfoundedness / ignorability). Implements three identification strategies: inverse probability weighting via the generalized propensity score, G-computation (outcome regression), and a doubly-robust combination. Supports both natural cubic splines and B-splines and provides pointwise confidence bands and fragility diagnostics. See Hirano and Imbens (2004) for background and the vignette for examples.

flexCausal v0.1.0: Provides doubly robust one-step and targeted maximum likelihood estimators for average causal effects in acyclic directed mixed graphs with unmeasured variables. Automatically determines whether the treatment effect is identified via backdoor adjustment or the extended front-door functional, and dispatches to the appropriate estimator. Supports incorporation of machine learning algorithms via SuperLearner and cross-fitting for nuisance estimation. Methods are described in Guo and Nabi (2024). See the vignette.

Computational Methods

opensimplex2 v0.0.3: Provides a function to generate 2, 3, or 4-dimensional gradient noise. The noise function is comparable to classic Perlin noise, but with less directional artefacts and lower computational overhead. It can have applications in procedural generation or flow field simulations. See the vignette.

ROOT v0.1.1: Implements a general framework for globally optimizing user-specified objective functionals over interpretable binary weight functions represented as sparse decision trees, called ROOT (Rashomon Set of Optimal Trees). It searches over candidate trees to construct a Rashomon set of near-optimal solutions and derives a summary tree highlighting stable patterns in the optimized weights. ROOT includes a built-in generalizability mode for identifying subgroups in trial settings for transportability analyses Parikh et al. (2025). There are three vignettes including Quickstart Guide and Optimization Path Example.

Data

edgarfundamentals v0.1.2: Provides a simple, ticker-based interface for retrieving fundamental financial data from the United States Securities and Exchange Commission’s EDGAR XBRL API. Functions return key financial ratios including earnings per share, return on equity, return on assets, debt-to-equity, current ratio, gross margin, operating margin, net margin, price-to-earnings, price-to-book, and dividend yield for any publicly traded U.S. company. Data is sourced directly from company 10-K annual filings, requiring no API key or paid subscription. Designed for use in quantitative finance courses and research workflows. See the vignette.

fred v0.2.0: Provides tidy access to economic data from the Federal Reserve Economic Data (FRED) API which is maintained by the Federal Reserve Bank of St. Louis and contains over 800,000 time series from 118 sources covering GDP, employment, inflation, interest rates, trade, and more. Functions include fetching series observations, search for series, browse categories, releases, and tags, and retrieve series metadata. Real-time and vintage helpers built on ALFRED return a series as it appeared on a given date, the first-release version, every revision, or a panel of selected vintages. Look here for an introduction.

readnoaa v0.1.1: Provides tidy access to climate and weather data from the National Oceanic and Atmospheric Administration (NOAA) via the National Centers for Environmental Information (NCEI) Data Service API. Covers daily weather observations, monthly and annual summaries, and 30-year climate normals from over 100,000 stations across 180 countries. No API key is required. Dedicated functions handle the most common datasets, while a generic fetcher provides access to all NCEI datasets. See README to get started.

Ecology

FuelDeep3D v0.1.1: Provides tools for preprocessing, feature extraction, and segmentation of three-dimensional forest point clouds derived from terrestrial laser scanning. Functions support creating height-above-ground metrics, tiling, and sampling point clouds, generating training datasets, applying trained models to new point clouds, and producing per-point fuel classes such as stems, branches, foliage, and surface fuels. Deep learning segmentation relies on the PointNeXt architecture described by Qian et al. (2022) while ground classification utilizes the Cloth Simulation Filter algorithm by Zhang et al. (2016). See README for examples.

pep725 v1.0.2: Provides a framework for quality-aware analysis of ground-based phenological data from the PEP725 Pan-European Phenology Database Templ et al. (2018) and Templ et al. (2026) and similar observation networks. Implements station-level data quality grading, outlier detection, phenological normals (climate baselines), anomaly detection, elevation and latitude gradient estimation with robust regression, spatial synchrony quantification, partial least squares regression for identifying temperature-sensitive periods, and sequential Mann-Kendall trend analysis. There are four vignettes including Getting Started and Phenological Analysis.

simpowa v1.0.3: Uses simulations from generalized linear mixed-effects models to incorporate random effects across multiple sources and levels of variation, and a dispersion parameter to account for overdispersion and capture unexplained variability. Covers design scenarios for both short-term and long-term trials evaluating the impact of single or combined vector control interventions. Methods build on Kipingu et al. (2025) and Johnson et al. (2015). See README to get started.

soilFlux v0.1.5: Implements a physics-informed one-dimensional convolutional neural network (CNN1D-PINN) for estimating the complete soil water retention curve (SWRC) as a continuous function of matric potential, from soil texture, organic carbon, bulk density, and depth. The network architecture ensures strict monotonic decrease of volumetric water content with increasing suction by construction, through cumulative integration of non-negative slope outputs (monotone integral architecture). Four physics-based residual constraints adapted from Norouzi et al. (2025). See README and the vignette.

Health Technology Assessment

drMAIC v0.1.0: Implements Doubly Robust Matching-Adjusted Indirect Comparison for population-adjusted indirect treatment comparisons in health technology appraisal. Features include standard MAIC via entropy balancing, augmented/doubly robust MAIC combining inverse probability weighting with outcome regression, comprehensive covariate balance diagnostics, Love plots, sensitivity analyses, bootstrap confidence intervals and submission-ready outputs aligned with NICE Decision Support Unit Technical Support Document 18, Cochrane Handbook guidance on indirect comparisons, and ISPOR best practice guidelines. See the vignette.

Mathematics

wkpool v0.3.0: Provides functions to establish and maintain vertex pool topology for geometry characterized by the Well-known text markup language. In this framework, segments are the atomic unit, vertices are shared via integer references into a pool. Topology is made discoverable via coincident vertex detection while not requiring modification of the input data. Topological data models follow principles described in Worboys and Duckham (2004). The edge-based topology geometry decomposed into vertices and directed edge pairs is a simplification of the quad-edge case in Guibas & Stolfi (1985). See the vignettes Segments: the atoms of geometry and wkpool: Topology-based Geometry Handling.

Medical Statistics

bmco v0.1.0: Provides Bayesian methods for comparing groups on multiple binary outcomes. Includes basic tests using multivariate Bernoulli distributions, subgroup analysis via generalized linear models, and multilevel models for clustered data. For statistical underpinnings, see Kavelaars, Mulder, and Kaptein (2020), Kavelaars, Mulder, and Kaptein (2024) and Kavelaars, Mulder, and Kaptein (2023). See the Shiny app for performing sample size computations. There are three vignettes including Introduction and Subgroup Analysis with Multivariate Binary Outcomes.

CohortMethod v6.0.1: Provides functions for performing comparative cohort studies in an observational database in the Observational Medical Outcomes Partnership (OMOP) Common Data Model which implements large-scale propensity scores (LSPS) as described in Tian et al. (2018). Functions are included for trimming, stratifying, matching and weighting by propensity scores, and diagnostics. See Suchard et al. (2013) for methods of using large scale regularized regression for fitting propensity models. There are three vignettes including Single Studies and Multiple Analyses.

epicR v1.0.1: Implements a Discrete Event Simulation model that simulates health outcomes of patients with Chronic Obstructive Pulmonary Disease (COPD) based on demographics and individual-level risk factors. It is based on the model published in Sadatsafavi et al. (2019). There are four vignettes including Getting Started and Background.

forestsearch v0.1.0: Designed for clinical researchers conducting exploratory subgroup analyses in randomized controlled trials, particularly for multi-regional clinical trials requiring regional consistency evaluation. The package implements statistical methods for exploratory subgroup identification in clinical trials with survival endpoints. Features include machine learning tools for identifying patient subgroups with differential treatment effects and bootstrap bias correction methods using the infinitesimal jackknife methods to address selection bias in post-hoc analyses. See León et al. (2024) for background and the vignette for examples.

mrddGlobal v0.1.0: Provides functions for global testing for regression discontinuity designs with more than one running variable including functions for testing whether there exist non-zero treatment effects along the boundary of the treated region, and testing whether there exist discontinuities in the joint density of the running variables along the boundary of the treated region. See Samiahulin (2026) for background and the vignette for examples.

Probability

ZeroOneDists v1.0.0: Implements new statistical distributions on the (0, 1) interval. Each distribution includes the traditional functions as well as an additional function called the family function, which can be used to estimate parameters using Generalized Additive Models for location, scale and shape, GAMLSS by Rigby & Stasinopoulos (2005).

Programming

arl v0.1.4: Provides a Scheme-inspired Lisp dialect embedded in R, with macros, tail-call optimization, and seamless interoperability with R functions and data structures. (The name ‘arl’ is short for ‘An R Lisp.’) Implemented in pure R with no compiled code. Note that the twenty-two vignettes only contain links to the source code. Look here to get started, and here for the layout of the source code.

restrictR v0.1.0: Provides functions to build reusable validators from small building blocks using the base pipe operator. Define runtime contracts once with restrict() and enforce them anywhere in code. Validators compose naturally, support dependent rules via formulas, and produce clear, path-aware error messages. No DSL, no operator overloading, just idiomatic R. See the vignette.

Public Health

episomer v3.0.34: Provides functions to automatically monitor trends of social media messages by time, place and topic aiming at detecting public health threats early through the detection of signals (i.e., an unusual increase in the number of messages per time, topic and location). It was designed to focus on infectious diseases, and it can be extended to all hazards or other fields of study by modifying the topics and keywords. More information on the original package epitweetr is available in the peer-review publication Espinosa et al. (2022). There are two vignettes: vignettes user documentation and add new social media. Look here for additional documentation including information about the associated Shiny app.

Risk Analysis

bayespmtools v0.0.1: Provides functions to perform Bayesian sample size, precision, and value-of-information analysis for external validation of existing multi-variable prediction models using the approach proposed by Sadatsafavi et al. (2025). There is a Getting Started Guide and a tutorial.

Statistics

badp v0.4.0.1: Implements Bayesian model averaging for dynamic panels with weakly exogenous regressors as described in the paper by Moral-Benito (2013) and provides functions to estimate dynamic panel data models and analyze the results of the estimation. See the vignette.

BORG v0.3.1: Provides tools for valid spatial, temporal, and grouped model evaluation. Automatically detects data dependencies (spatial autocorrelation, temporal structure, clustered observations), generates appropriate cross-validation schemes (spatial blocking, checkerboard, hexagonal, KNNDM, environmental blocking, leave-location-out, purged CV), and validates evaluation pipelines for leakage. Includes area of applicability assessment following Meyer & Pebesma (2021). Roberts et al. (2017) and Linnenbrink et al. (2024). There are three vignettes including Quick Start and Risk Taxonomy.

flexhaz v0.5.1: Implements a framework for specifying survival distributions through their hazard (failure rate) functions. Functions enable defining arbitrary time-varying hazard functions to model complex failure patterns including bathtub curves, proportional hazards with covariates, automatically compute survival, CDF, PDF, and quantile estimates, and implement the likelihood model interface for maximum likelihood estimation with right-censored and left-censored survival data. There are five vignettes including flexhaz: Hazard-First Survival Modeling and Dynamic Failure Rate Distributions.

maskedcauses v0.9.3: Implements maximum likelihood estimation for series systems where the component cause of failure is masked. Includes analytical log-likelihood, score, and Hessian functions for exponential, homogeneous Weibull, and heterogeneous Weibull component lifetimes under masked cause conditions. Supports exact, right-censored, left-censored, and interval-censored observations via composable observation functors. Also provides random data generation, model fitting, and Fisher information for asymptotic inference. See Lin, Loh, and Bai (1993) and Craiu and Reiser (2006) for background. There are seven vignettes including Censoring Types in Series System Masked Data and Model Selection for Masked Series Systems via Likelihood Ratio Tests.

mmbcv v0.3.0: Provides a function to compute robust and bias-corrected sandwich variance estimators for multi-state Cox models with clustered time-to-event data. The methodology extends the marginal Cox model bias-correction framework of Wang et al. (2023) to the multi-state setting. See the vignette.

osktnorm v1.1.2: Implements a moment-targeting normality transformation based on the simultaneous optimization of Tukey g-h distribution parameters designed to minimize both asymmetry (skewness) and excess peakedness (kurtosis) in non-normal data by mapping it to a standard normal distribution as described by Cebeci et al. (2026). Optimization is performed by minimizing an objective function derived from the Anderson-Darling goodness-of-fit statistic with Stephens’s correction factor, utilizing the L-BFGS-B algorithm for robust parameter estimation. This approach provides an effective alternative to power transformations like Box-Cox and Yeo-Johnson, particularly for data requiring precise tail-behavior adjustment. See the vignettes OSKT: Normality Transformation via Optimized Skewness and Kurtosis and A Practical Example for Working with Datasets.

RFmstate v0.1.2: Provides functions to fit cause-specific random survival forests for flexible multistate survival analysis with covariate-adjusted transition probabilities computed via product-integral. State transitions are modeled by random forests. Subject-specific transition probability matrices are assembled from predicted cumulative hazards using the product-integral formula. Also provides a standalone Aalen-Johansen nonparametric estimator as a covariate-free baseline, per-transition feature importance, bias-variance diagnostics, and comprehensive visualizations. Handles right censoring and competing transitions. Methods are described in Ishwaran et al. (2008) for random survival forests, Putter et al. (2007) for multistate competing risks decomposition, and Aalen and Johansen (1978) for the nonparametric estimator. See the vignette for an introduction.

SuperSurv v0.1.1: Implements a Super Learner framework for right-censored survival data. The package fits convex combinations of parametric, semiparametric, and machine learning survival learners by minimizing cross-validated risk using inverse probability of censoring weighting. It provides tools for automated hyperparameter grid search, high-dimensional variable screening, and evaluation of prediction performance using metrics such as the Brier score, Uno’s C-index, and time-dependent area under the curve. See Westling et al. (2024) and Lyu et al. (2026) for background. There are eleven vignettes including Machine Learning and 8. Interpreting the Black Box with SHAP & survex.

Time Series

betaARMA v1.1.0: Provides functions to fit Beta Autoregressive Moving Average (BARMA) models for time series data distributed in the standard unit interval (0, 1). The estimation is performed via the conditional maximum likelihood method using the Broyden-Fletcher-Goldfarb-Shanno (BFGS) quasi-Newton algorithm. The package includes tools for model fitting, diagnostic checking, and forecasting. Based on the work of Rocha and Cribari-Neto (2009) and the associated erratum Rocha and Cribari-Neto (2017). Look here to get started.

boundedur v1.0.1: Implements unit root tests for bounded time series following Cavaliere and Xu (2014). Standard unit root tests (ADF, Phillips-Perron) have non-standard limiting distributions when the time series is bounded. This package provides modified ADF and M-type tests (MZ-alpha, MZ-t, MSB) with p-values computed via Monte Carlo simulation of bounded Brownian motion. Supports one-sided (lower bound only) and two-sided bounds, with automatic lag selection using the MAIC criterion of Ng and Perron (2001). See README to get started.

reviser v0.1.1: Analyzes revisions in real-time time series vintages. Functions convert between wide revision triangles and tidy long vintages, extract selected releases, compute revision series, visualize vintage paths, and summarize revision properties such as bias, dispersion, autocorrelation, and news-noise diagnostics. There are also functions to identify efficient releases and estimate state-space models for revision nowcasting. Methods are based on Howrey (1978), Jacobs and Van Norden (2011), and Kishor and Koenig (2012). There are seven vignettes including Introduction and revision analysis.

Utilities

RDesk v1.0.5: Allows R developers to build native Windows desktop applications using R and WebView and create professional, standalone desktop tools with modern web-based user interfaces while maintaining a pure R backend. Provides a robust R6-based event loop, asynchronous background task management via mirai and callr, and a native Win32 message bridge for seamless R-to-user-interface communication without listening ports or network overhead. There are seven vignettes including Getting Started and Cookbook.

starburst v0.3.8: Implements a future backend that enables seamless execution of parallel R workloads on Amazon Web Services (‘AWS’), including EC2 and Fargate that handles environment synchronization, data transfer, quota management, and worker orchestration automatically, allowing users to scale from local execution to 100+ cloud workers with a single line of code change. There are twelve vignettes including Getting Started and Example: Parallel Portfolio Risk Modeling.

Visualization

BayesVolcano v1.0.1: Bayesian models are used to estimate effect sizes (e.g., gene expression changes, protein abundance differences, drug response effects) while accounting for uncertainty, small sample sizes, and complex experimental designs. However, Bayesian posteriors of models with many parameters are often difficult to interpret at a glance. One way to quickly identify important biological changes based on frequentist analysis is volcano plots (using fold-changes and p-values). Bayesian volcano plots bring together the explicit treatment of uncertainty in Bayesian models and the familiar visualization of volcano plots. See Sousa (2020) for background and the vignette for examples.

forrest v0.3.0: Provides a function to create publication-ready forest plots from any tabular data containing point estimates and confidence intervals. Suitable for visualizing results from regression models, meta-analyses, subgroup analyses, or any comparative study. Supports group and subgroup headings, summary estimates displayed as diamonds, grouped estimates with automatic color and shape mapping, vertical dodging of multiple estimates within the same row, customisable text columns alongside the plot, and optional row striping. Provides a helper to export plots to PDF, PNG, SVG, or TIFF. Built on tinyplot for clean, consistent visual styling with a minimal dependency footprint. There are four vignettes including Introduction to forrest and How forrest works.

HeatmapR v1.0.0: Provides a lightweight framework for creating high-quality, complex heatmaps using base graphics. Supports hierarchical clustering with dendrograms, column and row scaling, cluster sub-divisions, customizable cell colors, shapes and sizes, legends, and flexible layouts for arranging multiple heatmaps. Designed to return plot objects that can be easily arranged with other plots without sacrificing resolution. Methods for hierarchical clustering and distance computations are described in Murtagh and Contreras (2012). Dendrogram visualization methods are based on the ggdendro package by de Vries and Ripley (2020). See the vignette.

Artificial Intelligence

quallmer v0.3.0: Provides tools for AI-assisted qualitative data coding using large language models (‘LLMs’) via the ellmer package, supporting providers including OpenAI, Anthropic, Google, Azure, and local models via Ollama including built-in codebooks for common applications such as sentiment analysis and policy coding. Functions enable creating custom notebooks, support systematic replication across models and settings, compute inter-coder reliability statistics and validation metrics, and provide audit trails for documenting coding workflows following Lincoln and Guba’s (1985) framework for establishing trustworthiness in qualitative research. See the vignette to get started.

Biology

BioGSP v1.0.0: Implementation of Graph Signal Processing methods including Spectral Graph Wavelet Transform for analyzing spatial patterns in biological data and provides tools for multi-scale analysis of biology spatial signals, including forward and inverse transforms, energy analysis, and visualization functions tailored for biological applications. See Hammond, Vandergheynst, and Gribonval (2011) and Yao et al. (2024) for biological application examples, and the vignette to get started.

RuHere v1.0.1: Automatically flags common spatial errors in biological data using metadata, a six stage workflow, and functions that specifically integrate specialist-curated range information to identify geographic errors and introductions that often escape standard automated validation procedures. For details on the methodology see Trindade & Caron (2026). There are five vignettes including Reducing sampling bias and Ensuring spatial consistency.

Buddhism

tipitaka.critical v1.0.0: A lemmatized critical edition of the complete Pali Canon (Tipitaka), the canonical scripture of Theravadin Buddhism. Based on a five-witness collation of the Pali Text Society edition via GRETIL, SuttaCentral, the Vipassana Research Institute Chattha Sangayana edition, the Buddha Jayanti Tipitaka, and the Thai Royal Edition. All text is lemmatized using the Digital Pali Dictionary, grouping inflected forms by dictionary headword. Covers all three pitakas, Sutta, Vinaya, Abhidhamma, with 5,777 individual text units. For background on the collation method, see Zigmond (2026) and the vignette.

Climate Science

tidyextreme v1.00: Provides functions to calculate Expert Team on Climate Change Detection and Indices from daily or hourly temperature and precipitation data along with functions for flexible data handling. See the vignette.

Computational Methods

compositional.mle v2.0.0: Provides composable optimization strategies for maximum likelihood estimation. Solvers are first-class functions that combine via sequential chaining, parallel racing, and random restarts. Implements gradient ascent, Newton-Raphson, quasi-Newton (BFGS), and derivative-free methods with support for constrained optimization and tracing. Returns mle objects compatible with algebraic.mle for downstream analysis. Methods are based on Nocedal J, Wright SJ (2006). There are five vignettes including Getting Started and Theory and Intuition Behind Numerical MLE.

nabla v0.7.1: Enables exact automatic differentiation for R functions and provides a composable derivative operator D that computes gradients, Hessians, Jacobians, and arbitrary-order derivative tensors at machine precision. D(D(f)) gives Hessians, D(D(D(f))) gives third-order tensors for skewness of maximum likelihood estimators, and so on to any order. Works through any R code including loops, branches, and control flow. There are five vignettes including an Introduction and Gradient andHessian Computation.

rLifting v0.9.0: Performs Wavelet Lifting Transforms focusing on signal denoising and functional data analysis (FDA). Implements a hybrid architecture with a zero-allocation C++ core for high-performance processing. Features include unified offline (batch) denoising, causal, real-time filtering using a ring buffer engine, and adaptive recursive thresholding. There are five vignettes including an Introduction and Real-time signal smoothing.

setweaver v1.0.0: Create sets of variables based on a mutual information. In this context, a set is a collection of distinct elements (e.g., variables) that can also be treated as a single entity. Mutual information, quantifies the dependence between two variables by expressing how much information about one variable can be gained from observing the other. See the vignette.

Data

geobounds v0.1.1: Provides tools for downloading data from geoBoundaries. Several administration levels are available. See Runfola, D. et al. (2020) for background and the vignette for examples.

healthbR v0.2.0: Provides easy access to Brazilian public health data from multiple sources including VIGITEL (Surveillance of Risk Factors for Chronic Diseases by Telephone Survey), PNS (National Health Survey), PNAD Continua (Continuous National Household Sample Survey), POF (Household Budget Survey with food security and consumption data), Censo Demografico (population denominators), SIM (Mortality Information System), SINASC (Live Birth Information System) and several other repositories. Data is downloaded from the Brazilian Ministry of Health and is returned in tidy format following tidyverse conventions. There are seventeen vignettes including an Introduction and Chronic Disease Risk Factors.

nhanesdata v4.1.0: Instant access to harmonized National Health and Nutrition Examination Survey NHANES data spanning 1999-2023. Retrieve pre-processed datasets from reliable cloud storage with automatic type reconciliation and integrated search tools for variables and datasets. Simplifies NHANES data workflows by handling cycle management and maintaining data consistency across survey waves. There are four vignettes including Getting Started and Available NAHANES Datasets.

roadDB v1.1.0: Provides interface to the ROCEEH Out of Africa Database (ROAD), a comprehensive resource for archaeological, anthropological, paleoenvironmental and geographic data from Africa and Eurasia dating from 3,000,000 to 20,000 years BP. Users can retrieve data from the online database at different levels of detail and customize search requests. See README to get started.

Ecology

spacemodR v0.1.3: Provides tools for modeling food web transfer based on an initial ground raster. It provides a directed acyclic graph structure for a set of rasters representing the flow of elements (e.g., food, energy, contaminants). It also includes tools for working with dispersal algorithms, enabling the combination of flux data with population movement. See the tutorial.

Epidemiology

baselinenowcast v0.2.0: Provides nowcasting methods based on using empirical delay distributions and uncertainty from past performance as well as a baseline method for developers of new nowcasting methods. The package supports standard data frame inputs as well as the direct use of reporting triangles, and is compatible with epinowcast objects and accommodates a wide spectrum of reporting schedules, including mixed patterns of reference and reporting (daily-weekly, weekly-daily). For background see Johnson et al. (2026). There are five vignettes including Getting Started and Mathematical Methods.

multigroup.vaccine v0.1.1: Provides functions for modeling infectious disease dynamics in populations with multiple subgroups having different vaccination rates, transmission characteristics, and contact patterns. Enables calculating outbreak sizes, automatically fetching U.S. census data, and exploring vaccination scenarios with an interactive shiny dashboard. See Nguyen et al. (2024) and Duong et al. (2026) for background. There are four vignettes including Getting Started and Measles Age-Structured Model.

ViroReportR v1.0.4: Implements tools for reporting and forecasting viral respiratory infections, using case surveillance data. Report generation tools for short-term forecasts, and validation metrics for an arbitrary number of customizable respiratory viruses. Estimation of the effective reproduction number is based on the EpiEstim framework described in work by Cori et al. (2013). See the vignette.

Genomics

archipelago v0.1.0: Provides a graphical method for joint visualization of Variant Set Association Test results and individual variant association statistics. The Archipelago method assigns genomic coordinates to variant set statistics, allowing simultaneous display of variant-level and set-level signals in a unified plot supporting interpretation of both collective and individual variant contributions. For more see Lawless et al. (2026) and the vignette.

SVG v1.0.0: Implements a unified framework for detecting spatially variable genes in spatial transcriptomics data. SVG detection methods including MERINGUE (Moran’s I based spatial autocorrelation), Giotto binSpect (binary spatial enrichment test), SPARK-X (non-parametric kernel-based test), and nnSVG (nearest-neighbor Gaussian processes) which are described in Miller et al. (2021), Dries et al. (2021), Zhu et al. (2021), and Weber et al. (2023). See the vignette for examples.

Machine Learning

nadir v0.0.1: Provides a functional programming implementation of the super learner algorithm, van der Laan et al. (2007), with an emphasis on supporting the use of formulas to specify learners. Includes the ability to use random-effects specified in formulas e.g. (y ~ (age | strata) + …) and to construct new learners by passing a functions. See the vignette for basic examples.

Medical Applications

bfbin2arm v0.1.0: Provides tools to design and analyze two-arm binomial clinical (phase II) trials using Bayes factors. Implements Bayes factors for point-null and directional hypotheses, predictive densities under different hypotheses, and power and sample size calibration with optional frequentist type-I error and power. See the vignette.

neuromapr v0.2.1: Implements spatial null models and coordinate-space transformations for statistical comparison of brain maps, following the framework described in Markello et al. (2022). Provides variogram-matching surrogates, Moran spectral randomization, and spin-based permutation tests. Includes an R interface to the neuromaps annotation registry for browsing, downloading, and comparing brain map annotations from the Open Science Framework. There are five vignettes including Getting Started and Surface Geometry.

Physics

HaDeX2 v1.0.0: Process, analyze and visualize Hydrogen Deuterium eXchange monitored by Mass Spectrometry experiments (HDX-MS) via a new reproducible workflow for the analysis of the HDX-MS data that includes uncertainty propagation, data aggregation and visualization on 3D structure, functions for data exploration, quality control and generation of publication-quality figures and a companion Shiny application. There are eleven vignettes including Calculations and Data Visualization.

Statistics

BCFM v1.0.0: Implements the Bayesian Clustering Factor Models for simultaneous clustering and latent factor analysis of multivariate longitudinal data. The model accounts for within-cluster dependence through shared latent factors while allowing heterogeneity across clusters, enabling flexible covariance modeling in high-dimensional settings. The methodology is described in Shin, Ferreira, and Tegge (2018). See the vignette for examples.

clmstan v0.1.1: Provides functions to fit cumulative link models for ordinal categorical data using CmdStanR. Supports various link functions including logit, probit, cloglog, loglog, cauchit, and flexible parametric links such as Generalized Extreme Value, Asymmetric Exponential Power, and Symmetric Power. Methods are described in Agresti (2010), Wang and Dey (2011), and Naranjo, Perez, and Martin (2015). See the vignette to get started.

dtms v0.4.2: Implements discrete-time multistate models, several ways of estimating parametric and nonparametric multistate models, and an extensive set of Markov chain methods which use transition probabilities derived from the multistate model. See Schneider et al. (2024), Dudel (2021), Dudel & Myrskylä (2020), and van den Hout (2017) for background and README to get started.

GAReg v0.1.0: Provides a genetic algorithm framework for regression problems requiring discrete optimization over model spaces with unknown or varying dimension, where gradient-based methods and exhaustive enumeration are impractical. The computation is built on the GA engine of Scrucca (2017), and changepointGA engine from Li and Lu (2024). In challenging high-dimensional settings, functions enable efficient search and delivers near-optimal solutions. See the vignette.

mixpower v0.1.0: Implements a simulation-based toolkit for power and sample-size analysis for linear and generalized linear mixed-effects models (LMMs and GLMMs). Supports Gaussian, binomial, Poisson, and negative binomial families via lme4; Wald and likelihood-ratio tests; multi-parameter sensitivity grids; power curves and minimum sample-size solvers; parallel evaluation with deterministic seeds; and functions for reproducibility. Run time diagnostics include failure rate, singular-fit rate, effective N and publication-ready summary tables. There are five brief vignettes including an Introduction and Running simulations.

rblimp v1.0.: Provides an interface to Blimp software for Bayesian latent variable modeling, missing data analysis, and multiple imputation. The package generates Blimp syntax, executes Blimp models, and imports results back into R as structured objects with methods for visualization and analysis. See README to get started.

rareflow v0.1.0: Provides variational flow-based methods for modeling rare events using Kullback–Leibler divergence, normalizing flows, Girsanov change of measure, and Freidlin–Wentzell action functionals and tools for rare-event inference, minimum-action paths, and quasi-potential computation in stochastic dynamical systems. Methods are based on Rezende and Mohamed (2015), Girsanov (1960), and Freidlin and Wentzell (2012). See the vignette.

sshist v0.1.3: Implements the Shimazaki-Shinomoto method for optimizing the bin width of a histogram. This method minimizes the mean integrated squared error and features a C++ backend for high performance and shift-averaging to remove edge-position bias. Ideally suited for time-dependent rate estimation and identifying intrinsic data structures. Supports both 1D and 2D data distributions. See Shimazaki and Shinomoto (2007) for more details and the vignette for an introduction.

Surveys

heaping v0.1.0: Provides methods for correcting heaping (digit preference) in survey data at the individual record level. Age heaping, where respondents disproportionately report ages ending in 0 or 5, is a common phenomenon that can distort demographic analyses. Unlike traditional smoothing methods that only correct aggregated statistics, this package corrects individual values by replacing a calculated proportion of heaped observations with draws from fitted truncated distributions. See the vignette.

metasurvey v0.0.21: Provides a step-based pipeline for reproducible survey data processing, building on the survey package for complex sampling designs. Supports rotating panels with bootstrap replicate weights, and provides a recipe system for sharing and reproducing data transformation workflows across survey editions. There are thirteen vignettes including Getting Started and Survey design and Validation.

Time Series

mhpfilter v0.1.0: Implements the Modified Hodrick-Prescott Filter for decomposing macroeconomic time series into trend and cyclical components via efficient C++ routines. Unlike the standard HP filter, functions estimate series-specific lambda values that minimize the GCV criterion. See Choudhary, Hanif and Iqbal (2014), and Coe and McDermott (1997) for background. There is an Introduction and a vignette on Modified HP Filter Theory.

RegimeChange v0.1.1: Implements a unified framework for detecting regime changes (changepoints) in time series data that includes both frequentist and Bayesian methods for univariate and multivariate series with detection of changes in mean, variance, trend, and distributional properties. See Page (1954), Killick, Fearnhead, and Eckley (2012) for frequentist methods and Adams and MacKay (2007). for Bayesian methods. There are three vignettes including and Introduction and Bayesian Changepoint Detection.

Utilities

birddog v1.0.0: Provides a unified set of methods to detect scientific emergence and technological trajectories in academic papers and patents by combining citation network analysis with community detection and attribute extraction, also applying natural language processing and structural topic modeling to uncover the contents of research communities. Applications of the method include: Souza et al. (2022) and Maria et al. (2023). See README for the methodology and look here for an introduction and examples.

phinterval v1.0.0: Implements the phinterval vector class for representing time spans that may contain gaps (disjoint intervals) or be empty. This class generalizes the lubridate package’s interval class to support vectorized set operations (intersection, union, difference, complement) that always return a valid time span, even when disjoint or empty intervals are created. See the vignette.

Visualization

dtGAP v0.0.2: Provides supervised generalized association plots based on decision trees and enhances decision tree visualization by incorporating Generalized Association Plots through matrix-based visualizations including confusion matrix maps, decision tree matrix maps, and predicted class membership maps. See README.

ggInterval v0.2.4: Extends ggplot2 to visualize symbolic interval-valued data with various plots via more general and flexible input arguments, and provides a function to transform classical data into symbolic data using both clustering algorithms and customized methods. See the vignette.

nomiShape v1.0.1: Provides tools for visualizing and analyzing the shape of discrete nominal frequency distributions and introduces centered frequency plots, in which nominal categories are ordered from the most frequent category at the center toward less frequent categories on both sides, facilitating the detection of distributional patterns such as uniformity, dominance, symmetry, skewness, and long-tail behavior. In addition, the package supports Pareto charts for the study of dominance and cumulative frequency structure in nominal data. There are twelve vignettes including Visualizing and Analyzing Distributions of Nominal Variables and Pareto Plots for Nominal Distributions.

Artificial Intelligence

kuzco v0.1.0: Provides functions to make computer vision tasks approachable in R by leveraging Large Language Models, including fine-tuned prompts, boilerplate functions, and input/output helpers for common computer vision workflows, such as classifying and describing images. Functions are designed to take images as input and return structured data, helping users build practical applications with minimal code. There are four vignettes, including getting started and batch image processing.

Computational Methods

couplr v1.0.10: Functions designed for matching plots, sites, samples, or any pairwise optimization problem, solve optimal pairing and matching problems using linear assignment algorithms. Several algorithms are provided, including Hungarian method Kuhn (1955), the Jonker-Volgenant shortest path algorithm Jonker and Volgenant (1987), and the auction algorithm (Bertsekas (1988). There are six vignettes, including Quick Start and The Algorithm Collection.

hexify v0.3.10: Implements the ISEA discrete global grid system (Sahr, White and Kimerling (2003)). Includes a fast C++ core for projection and aperture quantization, and sf/terra-compatible R wrappers for grid generation and coordinate assignment. Output is compatible with dggridR for interoperability. There are three vignettes: Quick Start, Visualization, and Practical Workflows.

Data

dsBaseClient v6.3.5: Provides client-side base functions for the DataSHIELD software suite, which allows non-disclosive federated analysis on sensitive data. Functions have been designed to only share non-disclosive summary statistics, with built-in automated output checking based on statistical disclosure control with data sites setting the threshold values for the automated output checks. See README to get started.

mongolstats v0.1.1: Provides a tidyverse-friendly client for the National Statistics Office of Mongolia PXWeb API with helpers to discover tables, variables, and fetch statistical data. Also includes utilities to retrieve Mongolia administrative boundaries (ADM0-ADM2) as sf objects from open sources for mapping and spatial analysis. There are five vignettes, including Getting Started and Discovering Public Health Data.

read.abares v2.0.0: Provides functions to download and import agricultural data from the Australian Bureau of Agricultural and Resource Economics and Sciences ABARES and Australian Bureau of Statistics ABS. Data types serviced include spreadsheets, comma-separated value (CSV) files, geospatial data, including shape files and geotiffs, covering topics including broadacre crops, livestock, soil data, commodities, and more. See the vignettes, Setting Global Options and Working with spatial data.

Dynamical Systems

blvim v0.1.1: Provides functions to estimate Boltzmann–Lotka–Volterra (BLV) interaction model efficiently. Enables programmatic and graphical exploration of the solution space of BLV models when parameters are varied. See Wilson, A. (2008) for background and the vignettes: Systematic exploration of the BLV solution space and Theoretical Background.

Ecology

mrangr v1.0.1: Implements a mechanistic and spatially explicit simulator of metacommunities and extends the rangr by adding the ability to simulate multiple species interacting through an asymmetric matrix of pairwise relationships, allowing users to model all types of biotic interactions — competitive, facilitative, or neutral — within spatially explicit virtual environments. See the vignette.

Epidemiology

rsurvstat v0.1.4 Provides an interface to the SurvStat web service from the Robert Koch Institute, allowing downloads of disease time series stratified by pathogen type and subtype, age, and geography from notifiable disease reports in Germany. See the vignette.

Economics

emburden v0.6.1: Provides functions to calculate and analyze household energy burden using the Net Energy Return aggregation methodology. Functions support weighted statistical calculations across geographic and demographic cohorts, with utilities for formatting results into publication-ready tables. Methods are based on Scheier & Kittner (2022). There are three vignettes including Getting Started and Temporal Energy Burden Analysis.

Finance

cre.dcf v0.0.3: Provides utilities to build unlevered and levered discounted cash flow tables for commercial real estate assets. Functions generate bullet and amortising debt schedules, compute credit metrics such as debt coverage ratios, debt service coverage ratios, interest coverage ratios, debt yield ratios, and forward loan-to-value ratios based on net operating income. The toolkit evaluates refinancing feasibility under alternative market scenarios and supports end-to-end scenario execution. There are nine vignettes including Getting Started and Investment styles panorama.

OptimalBinningWoE v1.0.8 Implements 36 high-performance binning algorithms for Weight of Evidence transformation in credit scoring and risk modeling, including advanced methods such as Mixed Integer Linear Programming, Genetic Algorithms, Simulated Annealing, and Monotonic Regression. Features automatic method selection based on information value maximization, strict monotonicity enforcement, and efficient handling of large datasets. Fully integrated with the tidymodels ecosystem for building robust machine learning pipelines. Based on methods described in Siddiqi (2006) and Navas-Palencia (2020). The vignette demonstrates practical applications using real-world credit data.

“}

Genetics

bifrost v0.1.3: Implements methods for detecting and visualizing cladogenic shifts in multivariate trait data on phylogenies. Implements penalized-likelihood multivariate generalized least squares models, enabling analyses of high-dimensional trait datasets and large trees. Includes a greedy step-wise shift-search algorithm following approaches developed in Smith et al. (2023), Berv et al. (2024), and methods described in Clavel et al. (2019). See the vignette.

visPedigree v1.0.1: Provides tools for tidying, analyzing, and visualizing animal pedigrees by modeling pedigrees as directed acyclic graphs. This ensures robust loop detection, efficient generation assignment, and optimal sub-population splitting. Key features include standardizing pedigree formats, flexible ancestry tracing, and generating legible vector-based PDF graphs. A unique compaction algorithm enables the visualization of massive pedigrees by grouping full-sib families. There are three vignettes, including How to draw a pedigree and Calculation and visualization of relationship matrix.

Genomics

OmicNetR v0.1.1: Provides an end-to-end workflow for integrative analysis of two omics layers using sparse canonical correlation analysis, including sample alignment, feature selection, network edge construction, and visualization of gene-metabolite relationships. The underlying methods are based on penalized matrix decomposition and sparse CCA Witten, Tibshirani and Hastie (2009) with design principles inspired by multivariate integrative frameworks such as mixOmics Rohart et al. (2017). See the vignette.

High Performance Computing

futurize v0.1.0: Provides a straightforward path to scalable parallel computing via the future ecosystem. The futurize() function, which transpiles calls to sequential map-reduce functions, is combined with R’s native pipe operator to provide a 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, and crossmap. There is also support for a growing set of domain-specific packages, including boot, glmnet, mgcv, lme4, and tm. See README to get started. There are twelve vignettes, including Parallelize base-R apply functions and Parallelize purrr functions.

Mathematics

codyna v0.1.0: Perform analysis of complex dynamic systems with a focus on the temporal unfolding of patterns, changes, and state transitions in behavioral data. Supports both time series and sequence data and provides tools for the analysis and visualization of complexity, pattern identification, trends, regimes, sequence typology, as well as early warning signals. See the vignette.

EmpiricalDynamics v0.1.2: Implements a comprehensive toolkit for discovering differential and difference equations from empirical time series data using symbolic regression. The package implements a complete workflow from data preprocessing, including Total Variation Regularized differentiation for noisy economic data, visual exploration of dynamical structure, and symbolic equation discovery via genetic algorithms. Functions leverage a high-performance Julia backend, SymbolicRegression.jl to provide industrial-grade robustness, physics-informed constraints, and rigorous out-of-sample validation. Designed for economists, physicists, and researchers studying dynamical systems from observational data. See the vignette.

Machine Learning

multiRL v0.2.3: Provides a general purpose toolbox for implementing Rescorla-Wagner models in multi-armed bandit tasks. As the successor and functional extension of the binaryRL package, multiRL modularizes the Markov Decision Process (MDP) into six core components that enable constructing custom models via intuitive if-else syntax and define latent learning rules for agents. See Wilson & Collins (2019) and look here for an overview.

rCISSVAE v0.0.4: Implements the clustering-Informed Shared-Structure Variational Autoencoder, a deep learning framework for missing data imputation introduced in Khadem Charvadeh et al. (2025. The model accommodates all three types of missing data mechanisms: Missing Completely At Random, Missing At Random, and Missing Not At Random. There are seven vignettes, including a quick start guide and Handling Binary and Categorical Variables.

slideimp v0.5.4: Provides fast k-nearest neighbors (K-NN) and principal component analysis (PCA) imputation algorithms for missing values in high-dimensional numeric matrices, i.e., epigenetic data. For extremely high-dimensional data with ordered features, a sliding window approach for K-NN or PCA imputation is provided. Additional features include group-wise imputation (e.g., by chromosome), hyperparameter tuning with repeated cross-validation, multi-core parallelization, and optional subset imputation. See Josse and Husson (2016) for background and the vignette for an example.

SportMiner v0.1.0: Provides a toolkit for mining, analyzing, and visualizing scientific literature in sport science and includes functions for retrieving abstracts from Scopus, preprocessing text data, performing advanced topic modeling using Latent Dirichlet Allocation, Structural Topic Models, and Correlated Topic Models, and for creating publication-ready visualizations, including keyword co-occurrence networks and topic trends. See Blei et al. (2003), Roberts et al. (2014), and Blei and Lafferty (2007) for background. There are two vignettes: Getting Started and Text Mining and Topic Modeling for Sport Science Literature.

xplainfi v1.0.0: Provides a consistent interface for common feature importance methods as described in Ewald et al. (2024), including permutation feature importance, conditional and relative feature importance, leave one covariate out, and Shapley additive global importance as well as feature sampling mechanisms to support conditional importance methods. See the vignettes Getting Started, Feature Samplers, and Simulation Settings.

Medical Applications

autoFlagR v1.0.0: Provides automated data quality auditing using unsupervised machine learning and AI-driven anomaly detection for data quality assessment. Primarily designed for Electronic Health Records (EHR) data, with benchmarking capabilities for validation and publication. Methods based on Liu et al. (2008) and Breunig et al. (2000). There are three vignettes, including Getting Started and Healthcare Data Quality Example.

repfun v0.1.2: Provides functions to mimic the style of traditional reporting macros for clinical trials. The purpose is to generate tables, listings, and figures that support clinical research. This package is well-suited for firms or individuals who wish to incorporate R without changing their ways of working, as it follows a traditional clinical research workflow. Invoke functions (instead of macros) to summarize data and produce formatted reports. This package differs from others in that it includes tools (wrappers) for both analyzing and reporting data. There are twenty-seven vignettes, including Global Reporting Setup and SAS Type Variable Expansion.

Networks

flownet v0.1.2: Provides high-performance tools for transport modeling: network processing, route enumeration, and traffic assignment. The package implements the Path-Sized Logit model for traffic assignment Ben-Akiva and Bierlaire (1999), an efficient route enumeration algorithm, and provides powerful utility functions for (multimodal) network generation, consolidation/contraction, and/or simplification. See the vignette.

Statistics

bayesDiagnostics v0.1.0: Provides comprehensive tools for Bayesian model diagnostics and comparison, including prior sensitivity analysis, posterior predictive checks Gelman et al. (2013), advanced model comparison using Pareto-smoothed importance sampling leave-one-out cross-validation Vehtari et al. (2017), convergence diagnostics, and prior elicitation tools. Integrates with brms, rstan, and rstanarm packages. See README to get started and the vignette for an introduction.

gradLasso v0.1.1: Implements LASSO regression using gradient descent with support for Gaussian, Binomial, Negative Binomial, and Zero-Inflated Negative Binomial (ZINB) families. Features cross-validation for determining lambda, stability selection, and bootstrapping for confidence intervals. Methods described in Tibshirani (1996) and Meinshausen and Buhlmann (2010). Look here for a quick start and see the vignette for an introduction.

mfcurve v1.0.2: Implements multi-factor curve analysis for grouped data replicating and extending the functionality of the Stata mfcurve. See Krähmer (2023) and Simonsohn, Simmons, and Nelson (2020) for background. Functions for preprocessing, statistical testing, and visualization of results with confidence intervals are included. There is an Introduction.

NMAR v0.1.2: Implements methods to estimate finite-population parameters under nonresponse that are not missing at random. Incorporates auxiliary information and user-specified response models, and supports independent samples and complex survey designs via objects from the survey package. See Qin, Leung and Shao (2002) and Riddles, Kim and Im (2016) for background. There are five vignettes including exptilt nonparam theory and Empirical Likelihood.

pmrm v0.0.2: A progression model for repeated measures is a continuous-time nonlinear mixed-effects model for longitudinal clinical trials in progressive diseases. Unlike mixed models for repeated measures which estimate treatment effects as linear combinations of additive effects on the outcome scale, PMRMs characterize treatment effects in terms of the underlying disease trajectory yielding clinically interpretable quantities. See Raket (2022) and Kristensen (2016) for background. There are three vignettes: Models, Usage and Validation.

RSTr v1.1.4: Implements a Gibbs Sampler for Poisson or Binomial discrete spatial data for a variety of Spatiotemporal Conditional Autoregressive (CAR) models. Includes measures to prevent estimate over-smoothing through a restriction of model informativeness for select models. Also provides tools to load output and get median estimates. Methods are from Besag, York, and Mollié (1991), Gelfand and Vounatsou (2003), Quick et al. (2017), and Quick et al. (2021). There are twelve vignettes, including an Introduction and the CAR Models.

uniLasso v2.11: Fits a univariate-guided sparse regression (lasso) by a two-stage procedure. The first stage fits p-separate univariate models to the response. The second stage gives more weight to the more important univariate features, and preserves their signs. It returns an objects that inherit from class glmnet. See Chatterjee, Hastie and Tibshirani (2025) for details.

Time Series

rjd3toolkit v3.6.0: Implements an R interface to JDemetra+ 3.x R ecosystem of time series analysis software which provides functions to create outlier regressors, define calendar regressors, fit Unobserved Components AutoRegressive Integrated Moving Average (UCARIMA) models, to test the presence of trading days or seasonal effects, and also to set specifications in pre-adjustment and benchmarking when using rjd3x13 or rjd3tramoseats. See the JDemetra link above for details.

Utilities

automerge v0.3.1: Provides R bindings to the Automerge Conflict-free Replicated Data Type (CRDT) library, which enables automatic merging of concurrent changes without conflicts, making it ideal for distributed systems, collaborative applications, and offline-first architectures. See Kleppmann et al. (2019) for background. There are five vignettes, including Getting Started and Understanding CRDTs in Automerge.

h5lite v2.0.0.2: Implements an interface for the Hierarchical Data Format 5 HDF5 library that bundles the necessary system libraries to ensure easy installation on all platforms. Features smart defaults that automatically map R objects (vectors, matrices, data frames) to efficient HDF5 types, removing the need to manage low-level details like data spaces or property lists. There are nine vignettes, including Getting Started and Parallel Processing.

softwareRisk v0.1.0: Provides functions that leverage the network-like architecture of scientific models together with software quality metrics to identify chains of function calls that are more prone to generating and propagating errors. Functions operate on tbl_graph objects representing call dependencies between functions (callers and callees) and compute risk scores for individual functions and for paths (sequences of function calls) based on cyclomatic complexity, in-degree, and betweenness centrality. Supports variance-based uncertainty and sensitivity analyses after Puy et al. (2022) to assess how risk scores change under alternative risk definitions. See the vignette.

Visualization

ggguides v1.1.4: Extends ggplot2 by providing one-liner functions for common legend and guide operations in ggplot2. Simplifies legend positioning, styling, wrapping, and collection across multi-panel plots created with patchwork or cowplot. There are five vignettes, including Getting Started and Styling & Customization.

gglycan v0.0.3: Extends ggplot2 to plot glycans following the symbol nomenclature for glycans using a standardized visual representation of glycan structures. See the vignette.

ggsced v0.1.6: Extends ggplot2 to create publication-ready graphics with professional phase change lines, support for multiple baseline designs, and styling functions that follow Single-Case Experimental Design (SCED) visualization conventions. Key functions include adding phase change demarcation lines to existing plots and formatting axes with broken axis appearance, commonly used in single-case research. See the vignette.

ggskewboxplots V1.0.0: Extends ggplot2 for creating skewed boxplots using several statistical methods, including those of Kimber (1990), Hubert and Vandervieren (2008), Adil et al. (2015), Babura et al., (2017), and Walker et al. (2018). See the vignette.

vbracket v1.1.0: Extends ggplot2 by adding publication-quality custom legends with vertical brackets. Designed for displaying statistical comparisons between groups, commonly used in scientific publications for showing significance levels. Features include adaptive positioning, automatic bracket spacing for overlapping comparisons, font family inheritance, and support for asterisks, p-values, or custom labels. Look here for examples.

Agriculture

manureshed v0.1.2: Implements a framework for analyzing agricultural nutrient balances across multiple spatial scales with integration of wastewater treatment plant effluent loads for both nitrogen and phosphorus. Supports classification of spatial units as nutrient sources, sinks, or balanced areas based on agricultural surplus and deficit calculations. Built-in datasets include agricultural nutrient balance data from NuGIS; The Fertilizer Institute and Plant Nutrition Canada, 1987-2016) and U.S. Environmental Protection Agency wastewater discharge ECHO Discharge Monitoring Report (2007-2016). See Akanbi et al. (2025) for the methodology. There are seven vignettes, including Getting Started and Advanced Features.

Artificial Intelligence

gooseR v0.1.1: Integrates Goose AI capabilities including memory management, visualization enhancements, and workflow automation. Save R objects to Goose memory, apply Block branding to visualizations, and manage data science project workflows. There are six vignettes, including Getting Started and Code Review and Testing.

pairwiseLLM v1.1.0: Provides a unified framework for generating, submitting, and analyzing pairwise comparisons of writing quality using large language models (LLMs). The package supports live and/or batch evaluation workflows across multiple providers (OpenAI, Anthropic, Google Gemini, Together AI, and locally-hosted Ollama models). Results can be modeled using Bradley–Terry (1952) or Elo rating methods as described in Clark et al. (2018) to derive writing quality scores. For information on the method of pairwise comparisons, see Thurstone (1927) and Heldsinger & Humphry (2010). There are three vignettes, including Getting Started and Advanced.

Audio Analysis

ReVAMP v1.0.1: Provides an interface to the Vamp audio analysis plugin system developed by Queen Mary University of London’s Centre for Digital Music that enables loading and running Vamp plugins for various audio analysis tasks, including tempo detection, onset detection, spectral analysis, and audio feature extraction. Supports mono and stereo audio with automatic channel adaptation and domain conversion. There is an Introduction and a vignette on Managing Plugin Paths.

Causal Inference

caugi v1.0.0: Implements a simple interface to build, structure, and examine causal relationships. There are five vignettes, including Getting Started and Visualizing Causal Graphs.

Computational Methods

bigPLSR v0.7.2: Provides fast partial least squares (PLS) for dense and out-of-core data that is optimized for bigmemory-backed matrices with streamed cross-products and chunked BLAS. For details, see Bertrand and Maumy (2023), Bertrand and Maumy (2023b), and Dayal and MacGregor (1997). Features include kernel logistic PLS with C++-accelerated alternating iteratively reweighted least squares (IRLS) updates, streamed reproducing kernel Hilbert space (RKHS) solvers with reusable centering statistics, and bootstrap diagnostics with graphical summaries for coefficients, scores, and cross-validation workflows. There are fourteen vignettes, including Automatic Algorithm Selection and Visualizing PLS Fits.

Ecology

ambiR v0.1.1: Provides functions to calculate AZTI’s Marine Biotic Index - AMBI of benthic fauna species according to their sensitivity to pollution. The Shannon Diversity Index H’ and the Danish benthic fauna quality index DKI (Dansk Kvalitetsindeks) can also be calculated, as well as the multivariate M-AMBI index. See Borja et al. (2000) for background. There are four vignettes, including Get Started and The AMBI index.

SHARK4R v1.0.3: Provides functions to retrieve, process, analyze, and quality-control marine physical, chemical, and biological data. The main focus is on Swedish monitoring data available through the SHARK database, with additional API support for Nordic Microalgae, Dyntaxa, World Register of Marine Species WoRMS, AlgaeBase, OBIS xylookup web service, and Intergovernmental Oceanographic Commission (IOC) - UNESCO databases on harmful algae and toxins. There are five vignettes including Quality Control of SHARK4R Data and Retrieve Data From SHARK.

Econometrics

gvcAnalyzer v0.1.1: Provides tools for decomposing Global Value Chain (GVC) participation and value-added trade. It implements the frameworks proposed by Borin and Mancini (2023) for source-based and sink-based decompositions, and by Borin, Mancini, and Taglioni (2025) for tripartite and output-based GVC measures. There are three vignettes including an Introduction and Trade vs Output Perspectives.

Epidemiology

MetaRVM v1.0.1: Simulates respiratory virus epidemics using meta-population compartmental models. Following Fadikar et. al. (2025) it implements a stochastic SEIRD (Susceptible-Exposed-Infected-Recovered-Dead) framework with demographic stratification by age, race, and geographic zones that supports complex epidemiological scenarios including asymptomatic and presymptomatic transmission, hospitalization dynamics, vaccination schedules, and time-varying contact patterns via mixing matrices. There are four vignettes including Getting Started and Running a Simulation.

sitrep v0.4.0: Loads the complete sitrep ecosystem for applied epidemiology analysis. This package provides report templates and automatically loads companion packages, including epitabulate (for epidemiological tables), epidict (for data dictionaries), epikit (for epidemiological utilities), and apyramid (for age-sex pyramids). There are seven vignettes including Origin Story and Guide: Measles Outbreak report.

Genetics

EZbakR v0.1.0: A complete rewrite and reimagining of bakR, Vock et al. (2025), designed to support a wide array of analyses of nucleotide recoding RNA-seq datasets of any type, including TimeLapse-seq/SLAM-seq/TUC-seq, Start-TimeLapse-seq, TT-TimeLapse-seq, and subcellular NR-seq. It extends standard NR-seq standard NR-seq mutational modeling to support multi-label analyses, and implements an improved hierarchical model to better account for transcript-to-transcript variance in metabolic label incorporation, and also generalized dynamical systems modeling of NR-seq data to support analyses of premature mRNA processing and flow between subcellular compartments. There are seven vignettes, including Quickstart and Linear Modeling.

gpyramid v0.0.1: Provides functions to identify efficient crossing schemes for gene pyramiding and calculates the cost of crossing in terms of the number of individuals and generations, which has been theoretically formulated by Servin et al. (2004). The package has been designed for selecting appropriate parental genotypes and finding the most efficient crossing scheme for gene pyramiding, especially for plant breeding. There is an Introduction and a Tutorial.

phylospatial v1.2.1: Provides functions to analyze spatial phylogenetic diversity patterns. Use your data on an evolutionary tree and geographic distributions of the terminal taxa to compute diversity and endemism metrics, test significance with null model randomization, analyze community turnover and biotic regionalization, and perform spatial conservation prioritizations. All functions support quantitative community data in addition to binary data. There are four vignettes, including phylospatial data and Prioritization.

Genomics

CimpleG v1.0.1: Provides a method for the detection of small CpG methylation signatures used for cell-type classification and deconvolution that is time efficient and performs as well as top performing methods for cell-type classification of blood cells and other somatic cells. Predictions are based on a single DNA methylation site per cell type, but users can also select more sites if they so wish. Users can train cell type classifiers and directly apply these in a deconvolution of cell mixes context. See Maié et al. (2023) for details and the vignettes Genetic signatures and Quickly save and load large objects.

Machine Learning

LBBNN v0.1.4: Implements an interface to the LibTorch backend to provide latent binary Bayesian neural networks via torch. Supports mean-field variational inference as well as flexible variational posteriors using normalizing flows. The standard LBBNN implementation follows Hubin and Storvik (2024) and uses the local reparametrization trick as in Skaaret-Lund et al. (2024). Input-skip connections are implemented as described in Høyheim et al. (2025). See the vignette to get started.

snic v0.6.1: Implements the Simple Non-Iterative Clustering algorithm for superpixel segmentation of multi-band images, as introduced by Achanta and Susstrunk (2017), and supports both standard image arrays and geospatial raster objects, with a design that can be extended to other spatial data frameworks. The algorithm groups adjacent pixels into compact, coherent regions based on spectral similarity and spatial proximity. A high-performance implementation supports images with arbitrary spectral bands. See the vignettes about Arrays and SpatRaster segmentation pipelines.

Mathematics

algebraic.dist v0.1.0: Provides an algebra over probability distributions enabling composition, sampling, and automatic simplification to closed forms. Supports normal, exponential, multivariate normal, and empirical distributions with operations like addition and subtraction that automatically simplify when mathematical identities apply (e.g., the sum of independent normal distributions is normal).

Mediation Analysis

wsMed v1.0.2: Provides tools for within-subject mediation analysis using structural equation modeling to examine how changes in an outcome variable between two conditions are mediated through one or more variables. Supports within-subject mediation analysis using the lavaan package by Rosseel (2012), and extends Monte Carlo confidence interval estimation to missing data scenarios using the semmcci package by Pesigan and Cheung (2023). There are six vignettes, including wsMed and GenerateModelIP.

Medical Statistics

cifmodeling v0.9.8: Implements a publication-ready toolkit for modern survival and competing risks analysis with a minimal, formula-based interface. Both nonparametric estimation and direct polytomous regression of cumulative incidence functions (CIFs) are supported. Functions estimate survival and CIF curves, produce high-quality graphics with risk tables, censoring and competing-risk marks, and multi-panel or inset layouts. All core functions adopt a formula-and-data syntax and return tidy and extensible outputs. Key numerical routines are implemented in C++. There are seven vignettes including Overview and Examples.

Medical Statistics Continued

eVCGsampler v0.9.2: Provides a principled framework for sampling Virtual Control Group (VCG) using energy distance-based covariate balancing, offers visualization tools to assess covariate balance, and includes a permutation test to evaluate the statistical significance of observed deviations. See the vignette.

fgdiR v0.1.0: provides tools for computing the Functional Gait Deviation Index, a novel index for quantifying gait pathology using multivariate functional principal component analysis and supports analysis at the level of both legs combined, individual legs, and individual joints/planes. It includes functions for functional data preprocessing, multivariate functional principal component decomposition, FGDI computation, and visualization of gait abnormality scores. See Minhas et al. (2025) for background and the vignettes Amputee Gait Analysis and Parkinson Gait Analysis.

Pharma

aNCA v0.1.0: Provides an interactive shiny application for performing non-compartmental analysis on pre-clinical and clinical pharmacokinetic data and provides interactive visualizations, CDISC outputs (ADNCA, PP, ADPP) and configurable TLGs (tables, listings, and graphs). Methods and core estimators are described in Denney, Duvvuri, and Buckeridge (2015). There are three vignettes, including a User Guide and Adding new TLGs.

httkexamples v0.0.1: Provide examples as vignettes using High throughput toxicokinetics (HTTK) methods to solve various problems in bioinformatics, toxicology, and exposure science. In accordance with Davidson-Fritz et al. (2025), whenever a new HTTK model is developed, the code to generate the figures evaluating that model is added as a new vignette. HTTKis the combination of chemical-specific in vitro measurements or in silico predictions and generic mathematical models, to predict absorption, distribution, metabolism, and excretion by the body. See Pearce et al. (2017) and Breen et al. (2021) for background. There are thirteen vignettes including HTTK Basics and Generating subpopulations.

Statistics

distionary v0.1.0: Provides functions to create and evaluate probability distribution objects from a variety of families or define custom distributions and automatically compute distributional properties, even when they have not been specified. This package supports statistical modeling and simulations, and forms the core of the probaverse suite of R packages. There are three vignettes, including Evaluate a Distribution and Specifying Your Own Distribution.

jollofR v0.6.5: Provides functions to automatically disaggregate small-area population estimates by demographic groups (e.g., age, sex, race, marital status, educational level, etc) along with the estimates of uncertainty, using advanced Bayesian statistical modelling approaches based on integrated nested Laplace approximation (INLA) Rue et al. (2009), and stochastic partial differential equation methods Lindgren et al. (2011), hierarchical Bayesian modeling frameworks for small area estimation as described in Leasure et al. (2020), and Nnanatu et al. (2025). See README for package details and examples.

PSsurvival v0.2.0: Implements propensity score weighting methods for estimating counterfactual survival functions, marginal hazard ratios, and weighted Kaplan-Meier and cumulative risk curves in observational studies with time-to-event outcomes. Supports binary and multiple treatment groups with inverse probability of treatment weighting, overlap weighting, and average treatment effect on the treated, and includes symmetric trimming (Crump extension) for extreme propensity scores. Implements variance estimation via analytical M-estimation or bootstrap. Methods based on Li et al. (2018), Li & Li (2019), and Cheng et al. (2022). See the vignette.

quantbayes v0.1.0: Implements the Quantification Evidence Standard algorithm for computing Bayesian evidence sufficiency from binary evidence matrices and provides posterior estimates, credible intervals, percentiles, and optional visual summaries. The method is universal, reproducible, and independent of any specific clinical or rule-based framework. See The Quantitative Omics Epidemiology Group et al. (2025) for details and the vignette for examples.

tipse v1.2: Implements tipping point sensitivity analysis for time-to-event endpoints under different missing data scenarios, as described in Oodally et al. (2025). Supports both model-based and model-free imputation, multiple imputation workflows, plausibility assessment, and visualizations. Enables robust assessment for regulatory and exploratory analyses. See the vignette.

weightedsurv v0.1.0: Provides survival analysis functions with support for time-dependent and subject-specific (e.g., propensity score) weighting. Implements weighted estimation for Cox models, Kaplan-Meier survival curves, and treatment differences with point-wise and simultaneous confidence bands. Includes restricted mean survival time comparisons evaluated across all potential truncation times with both point-wise and simultaneous confidence bands. See Cole & Hernán (2004) for methodological background and the vignette for examples.

Time Series

bpvars v1.0: Provides Bayesian estimation and forecasting of dynamic panel data using Bayesian Panel Vector Autoregressions with hierarchical prior distributions. Models include country-specific VARs that share a global prior distribution that extends the model by Jarociński (2010). Functions provide model specification, estimation, and forecasting routines, facilitating coherent workflows and reproducibility and includes automated pseudo-out-of-sample forecasting and computation of forecasting performance measures and algorithms written in C++. See README to get started.

gctsc V0.1.3: Implements Gaussian copula models for count time series. Includes simulation utilities, likelihood approximation, maximum-likelihood estimation, residual diagnostics, and predictive inference. Implements the Time Series Minimax Exponential Tilting method, an adaptation of Minimax Exponential Tilting Botev, 2017, and the Vecchia-based tilting framework of Cao and Katzfuss (2025). Also provides a linear-cost implementation of the Geweke–Hajivassiliou–Keane simulator inspired by Masarotto and Varin (2012), and the Continuous Extension approximation of Nguyen and De Oliveira (2025). See the vignette.

Utilities

fjoin v0.1.0: Extends data.table join functionality to work with any data frame class, and provides a familiar x/y-style interface, enabling broad use across R. Offers NA-safe matching by default, on-the-fly column selection, multiple match-handling on both sides, x or y row order, and a row origin indicator. Performs inner, left, right, full, semi- and anti-joins with equality and inequality conditions, plus cross joins. Specific support for data.table, (grouped) tibble, and sf/sfc objects and their attributes; returns a plain data frame otherwise. Avoids data-copying of inputs and outputs. Allows displaying the data.table code instead of (or as well as) executing it. See the vignette.

meetupr v0.3.1: Provides programmatic access to the Meetup GraphQL API enabling users to retrieve information about groups, events, and members from Meetup. Supports authentication via OAuth2 and includes functions for common queries and data manipulation tasks. There are four vignettes, including Getting Started and Meetup API Schema Introspection.

odiffr v0.5.1: Implements R bindings to odiff, a pixel-by-pixel image comparison tool. Supports PNG, JPEG, WEBP, and TIFF with configurable thresholds, antialiasing detection, and region ignoring. Requires system installation of odiff. Ideal for visual regression testing in automated workflows. See the vignette.

svgedit v1.0.0: Provides functions to edit SVG files created in Inkscape by replacing placeholders (e.g., a rectangle element or {} in a text box by ggplot2 objects, images, or text. This helps automate the creation of figures with complex layouts. See the vignette.

taskqueue v0.2.0: Implements a task queue system for asynchronous parallel computing using PostgreSQL as a backend. Designed for embarrassingly parallel problems where tasks do not communicate with each other. Dynamically distributes tasks to workers, handles uneven load balancing, and allows new workers to join at any time. Particularly useful for running large numbers of independent tasks on high-performance computing (HPC) clusters with SLURM https://slurm.schedmd.com/ job schedulers.

yaml12 v0.1.0: Provides a fast, correct, safe, and ergonomic YAML 1.2 parser and generator written in Rust, enabling conversion between YAML and simple R objects with full support for multi-document streams, tags, anchors, and aliases. Offers opt-in handlers for custom tag behavior and round-trips common R data structures. Implements the YAML 1.2.2 specification. There are two vignettes, YAML in 2 minutes and YAML Tags.

Visualization

blockr.ggplot v0.1.0: Extends blockr.core with interactive blocks for data visualization using ggplot2. Users can build charts through a graphical interface without writing code directly. Includes common chart types (bar charts, line charts, pie charts, scatter plots) as well as statistical plots (boxplots, histograms, density plots, violin plots) with rich customization options and intuitive user interfaces. See the vignette.

mapycusmaximus v1.0.0: Provides focus-glue-context fisheye transformations to two-dimensional coordinates and spatial vector geometries. Implements a smooth radial distortion that enlarges a focal region, transitions through a glue ring, and preserves outside context. Methods build on generalized fisheye views and focus+context mapping. See Furnas (1986), Furnas (2006), and Yamamoto et al. (2009) for details and the vignette for examples.