This package implements collaborative filtering methods for recommendation systems based on user-item interaction data. Supports both explicit feedback (ratings) and implicit feedback (consumption). The package uses efficient sparse matrix representations and provides incremental updates for users, items, and similarity structures through an R6 class-based architecture. See Aggarwal (2016) <doi:10.1007/978-3-319-29659-3> for an overview.

This package provides tools for penalized estimation of flexible hidden Markov models for time series of counts w/o the need to specify a (parametric) family of distributions. These include functions for model fitting, model checking, and state decoding. For details, see Adam, T., Langrock, R., and Weià , C.H. (2019): Penalized Estimation of Flexible Hidden Markov Models for Time Series of Counts. <arXiv:1901.03275>.

This package provides the tools to produce catseye plots, principally by catseyesplot() function which calls R's standard plot() function internally, or alternatively by the catseyes() function to overlay the catseye plot onto an existing R plot window. Catseye plots illustrate the normal distribution of the mean (picture a normal bell curve reflected over its base and rotated 90 degrees), with a shaded confidence interval; they are an intuitive way of illustrating and comparing normally distributed estimates, and are arguably a superior alternative to standard confidence intervals, since they show the full distribution rather than fixed quantile bounds. The catseyesplot and catseyes functions require pre-calculated means and standard errors (or standard deviations), provided as numeric vectors; this allows the flexibility of obtaining this information from a variety of sources, such as direct calculation or prediction from a model. Catseye plots, as illustrations of the normal distribution of the means, are described in Cumming (2013 & 2014). Cumming, G. (2013). The new statistics: Why and how. Psychological Science, 27, 7-29. <doi:10.1177/0956797613504966> pmid:24220629.

Fit composite Gaussian process (CGP) models as described in Ba and Joseph (2012) "Composite Gaussian Process Models for Emulating Expensive Functions", Annals of Applied Statistics. The CGP model is capable of approximating complex surfaces that are not second-order stationary. Important functions in this package are CGP, print.CGP, summary.CGP, predict.CGP and plotCGP.

Parameters of a user-specified probability distribution are modelled by a multi-layer perceptron artificial neural network. This framework can be used to implement probabilistic nonlinear models including mixture density networks, heteroscedastic regression models, zero-inflated models, etc. following Cannon (2012) <doi:10.1016/j.cageo.2011.08.023>.

Load Current Population Survey (CPS) microdata into R using the Census Bureau Data API (<https://www.census.gov/data/developers/data-sets.html>), including basic monthly CPS and CPS ASEC microdata.

This package provides a collection of functions to calculate Composite Indicators methods, focusing, in particular, on the normalisation and weighting-aggregation steps, as described in OECD Handbook on constructing composite indicators: methodology and user guide, 2008, Vidoli and Fusco and Mazziotta <doi:10.1007/s11205-014-0710-y>, Mazziotta and Pareto (2016) <doi:10.1007/s11205-015-0998-2>, Van Puyenbroeck and Rogge <doi:10.1016/j.ejor.2016.07.038> and other authors.

Estimation of average treatment effects (ATE) of point interventions on time-to-event outcomes with K competing risks (K can be 1). The method uses propensity scores and inverse probability weighting for emulation of baseline randomization, which is described in Charpignon et al. (2022) <doi:10.1038/s41467-022-35157-w>.

Imports and cleans opencovid19-fr <https://github.com/opencovid19-fr/data> data on COVID-19 in France.

Reading and writing of files in the most commonly used formats of structural crystallography. It includes functions to work with a variety of statistics used in this field and functions to perform basic crystallographic computing. References: D. G. Waterman, J. Foadi, G. Evans (2011) <doi:10.1107/S0108767311084303>.

Stan based functions to estimate CAR-MM models. These models allow to estimate Generalised Linear Models with CAR (conditional autoregressive) spatial random effects for spatially and temporally misaligned data, provided a suitable Multiple Membership matrix. The main references are Gramatica, Liverani and Congdon (2023) <doi:10.1214/23-BA1370>, Petrof, Neyens, Nuyts, Nackaerts, Nemery and Faes (2020) <doi:10.1002/sim.8697> and Gramatica, Congdon and Liverani <doi:10.1111/rssc.12480>.

Connect to WFP's Moda platform to R, download data, and obtain the list of individuals with access to the project along with their access level.

This package performs copy number variants association analysis with Lasso and Weighted Fusion penalized regression. Creates a "CNV profile curve" to represent an individualâ s CNV events across a genomic region so to capture variations in CNV length and dosage. When evaluating association, the CNV profile curve is directly used as a predictor in the regression model, avoiding the need to predefine CNV loci. CNV profile regression estimates CNV effects at each genome position, making the results comparable across different studies. The penalization encourages sparsity in variable selection with a Lasso penalty and encourages effect smoothness between consecutive CNV events with a weighted fusion penalty, where the weight controls the level of smoothing between adjacent CNVs. For more details, see Si (2024) <doi:10.1101/2024.11.23.624994>.

Flexible framework for trait-based simulation of community assembly, where components could be replaced by user-defined function and that allows variation of traits within species.

Linear or nonlinear cross-lagged panel model can be built from input data. Users can choose the appropriate method from three methods for constructing nonlinear cross lagged models. These three methods include polynomial regression, generalized additive model and generalized linear mixed model.In addition, a function for determining linear relationships is provided. Relevant knowledge of cross lagged models can be learned through the paper by Fredrik Falkenström (2024) <doi:10.1016/j.cpr.2024.102435> and the paper by A Gasparrini (2010) <doi:10.1002/sim.3940>.

Cox model inference for relative hazard and covariate-specific pure risk estimated from stratified and unstratified case-cohort data as described in Etievant, L., Gail, M.H. (Lifetime Data Analysis, 2024) <doi:10.1007/s10985-024-09621-2>.

Routines doing cone projection and quadratic programming, as well as doing estimation and inference for constrained parametric regression and shape-restricted regression problems. See Mary C. Meyer (2013)<doi:10.1080/03610918.2012.659820> for more details.

Computes a range of scatterplot diagnostics (scagnostics) on pairs of numerical variables in a data set. A range of scagnostics, including graph and association-based scagnostics described by Leland Wilkinson and Graham Wills (2008) <doi:10.1198/106186008X320465> and association-based scagnostics described by Katrin Grimm (2016,ISBN:978-3-8439-3092-5) can be computed. Summary and plotting functions are provided.

We present corto (Correlation Tool), a simple package to infer gene regulatory networks and visualize master regulators from gene expression data using DPI (Data Processing Inequality) and bootstrapping to recover edges. An initial step is performed to calculate all significant edges between a list of source nodes (centroids) and target genes. Then all triplets containing two centroids and one target are tested in a DPI step which removes edges. A bootstrapping process then calculates the robustness of the network, eventually re-adding edges previously removed by DPI. The algorithm has been optimized to run outside a computing cluster, using a fast correlation implementation. The package finally provides functions to calculate network enrichment analysis from RNA-Seq and ATAC-Seq signatures as described in the article by Giorgi lab (2020) <doi:10.1093/bioinformatics/btaa223>.

Predicts anticancer peptides using random forests trained on the n-gram encoded peptides. The implemented algorithm can be accessed from both the command line and shiny-based GUI. The CancerGram model is too large for CRAN and it has to be downloaded separately from the repository: <https://github.com/BioGenies/CancerGramModel>. For more information see: Burdukiewicz et al. (2020) <doi:10.3390/pharmaceutics12111045>.

Set of forecasting tools to predict ICU beds using a Vector Error Correction model with a single cointegrating vector. Method described in Berta, P. Lovaglio, P.G. Paruolo, P. Verzillo, S., 2020. "Real Time Forecasting of Covid-19 Intensive Care Units demand" Health, Econometrics and Data Group (HEDG) Working Papers 20/16, HEDG, Department of Economics, University of York, <https://www.york.ac.uk/media/economics/documents/hedg/workingpapers/2020/2016.pdf>.

This package provides tools for causal structure learning from observational data, with emphasis on temporally ordered variables. The package implements the Temporal Peterâ Clark (TPC) algorithm (Petersen, Osler & Ekstrøm, 2021; <doi:10.1093/aje/kwab087>), the Temporal Greedy Equivalence Search (TGES) algorithm (Larsen, Ekstrøm & Petersen, 2025; <doi:10.48550/arXiv.2502.06232>) and Temporal Fast Causal Inference (TFCI). It provides a unified framework for specifying background knowledge, which can be incorporated into the implemented algorithms from the R packages bnlearn (Scutari, 2010; <doi:10.18637/jss.v035.i03>) and pcalg (Kalish et al., 2012; <doi:10.18637/jss.v047.i11>), as well as the Java library Tetrad (Scheines et al., 1998; <doi:10.1207/s15327906mbr3301_3>). The package further includes utilities for visualization, comparison, and evaluation of graph structures, facilitating performance evaluation and methodological studies.

Colorful Data Frames in the terminal. The new class does change the behaviour of any of the objects, but adds a style definition and a print method. Using ANSI escape codes, it colors the terminal output of data frames. Some column types (such as p-values and identifiers) are automatically recognized.

Providing more beautiful and more meaningful return messages for checkmate assertions and checks helping users to better understand errors.