This package performs repeated nested cross-validation for Cox Proportionate Hazards, Cox Lasso, Survival Random Forest, and their ensemble. Returns internally validated concordance index, time-dependent area under the curve, Brier score, calibration slope, and statistical testing of non-linear ensemble outperforming the baseline Cox model. In this, it helps researchers to quantify the gain of using a more complex survival model, or justify its redundancy. Equally, it shows the performance value of the non-linear and interaction terms, and may highlight the need of further feature transformation. Further details can be found in Shamsutdinova, Stamate, Roberts, & Stahl (2022) "Combining Cox Model and Tree-Based Algorithms to Boost Performance and Preserve Interpretability for Health Outcomes" <doi:10.1007/978-3-031-08337-2_15>, where the method is described as Ensemble 1.

This package provides robust and efficient methods for estimating causal effects in a target population using a multi-source dataset, including those of Dahabreh et al. (2019) <doi:10.1111/biom.13716>, Robertson et al. (2021) <doi:10.48550/arXiv.2104.05905>, and Wang et al. (2024) <doi:10.48550/arXiv.2402.02684>. The multi-source data can be a collection of trials, observational studies, or a combination of both, which have the same data structure (outcome, treatment, and covariates). The target population can be based on an internal dataset or an external dataset where only covariate information is available. The causal estimands available are average treatment effects and subgroup treatment effects. See Wang et al. (2025) <doi:10.1017/rsm.2025.5> for a detailed guide on using the package.

This package provides functions for evaluating the stability of low-dimensional embeddings and cluster assignments in singleâ cell RNA sequencing (scRNAâ seq) datasets. Starting from a principal component analysis (PCA) object, users can generate multiple replicates of tâ Distributed Stochastic Neighbor Embedding (tâ SNE) or Uniform Manifold Approximation and Projection (UMAP) embeddings. Embedding stability is quantified by computing pairwise Kendallâ s Tau correlations across replicates and summarizing the distribution of correlation coefficients. In addition to dimensionality reduction, scStability assesses clustering consistency using either Louvain or Leiden algorithms and calculating the Normalized Mutual Information (NMI) between all pairs of cluster assignments. For background on UMAP and t-SNE algorithms, see McInnes et al. (2020, <doi:10.21105/joss.00861>) and van der Maaten & Hinton (2008, <https://github.com/lvdmaaten/bhtsne>), respectively.

This package provides a framework for automated machine learning. Concretely, the focus is on the optimisation of bagging workflows. A bagging workflows is composed by three phases: (i) generation: which and how many predictive models to learn; (ii) pruning: after learning a set of models, the worst ones are cut off from the ensemble; and (iii) integration: how the models are combined for predicting a new observation. autoBagging optimises these processes by combining metalearning and a learning to rank approach to learn from metadata. It automatically ranks 63 bagging workflows by exploiting past performance and dataset characterization. A complete description of the method can be found in: Pinto, F., Cerqueira, V., Soares, C., Mendes-Moreira, J. (2017): "autoBagging: Learning to Rank Bagging Workflows with Metalearning" arXiv preprint arXiv:1706.09367.

Extends the Seurat classes and functions to support Genomic Data Structure (GDS) files as a DelayedArray backend for data representation. It relies on the implementation of GDS-based DelayedMatrix in the SCArray package to represent single cell RNA-seq data. The common optimized algorithms leveraging GDS-based and single cell-specific DelayedMatrix (SC_GDSMatrix) are implemented in the SCArray package. SCArray.sat introduces a new SCArrayAssay class (derived from the Seurat Assay), which wraps raw counts, normalized expressions and scaled data matrix based on GDS-specific DelayedMatrix. It is designed to integrate seamlessly with the Seurat package to provide common data analysis in the SeuratObject-based workflow. Compared with Seurat, SCArray.sat significantly reduces the memory usage without downsampling and can be applied to very large datasets.

This package provides a (mildly) opinionated set of functions to help assess medication adherence for researchers working with medication claims data. Medication adherence analyses have several complex steps that are often convoluted and can be time-intensive. The focus is to create a set of functions using "tidy principles" geared towards transparency, speed, and flexibility while working with adherence metrics. All functions perform exactly one task with an intuitive name so that a researcher can handle details (often achieved with vectorized solutions) while we handle non-vectorized tasks common to most adherence calculations such as adjusting fill dates and determining episodes of care. The methodologies in referenced in this package come from Canfield SL, et al (2019) "Navigating the Wild West of Medication Adherence Reporting in Specialty Pharmacy" <doi:10.18553/jmcp.2019.25.10.1073>.

This package provides a collection of functions for exploratory chemometrics of 2D spectroscopic data sets such as COSY (correlated spectroscopy) and HSQC (heteronuclear single quantum coherence) 2D NMR (nuclear magnetic resonance) spectra. ChemoSpec2D deploys methods aimed primarily at classification of samples and the identification of spectral features which are important in distinguishing samples from each other. Each 2D spectrum (a matrix) is treated as the unit of observation, and thus the physical sample in the spectrometer corresponds to the sample from a statistical perspective. In addition to chemometric tools, a few tools are provided for plotting 2D spectra, but these are not intended to replace the functionality typically available on the spectrometer. ChemoSpec2D takes many of its cues from ChemoSpec and tries to create consistent graphical output and to be very user friendly.

This package provides functions to calculate step- and cadence-based metrics from timestamped accelerometer and wearable device data. Supports CSV and AGD files from ActiGraph devices, CSV files from Fitbit devices, and step counts derived with R package GGIR <https://github.com/wadpac/GGIR>, with automatic handling of epoch lengths from 1 to 60 seconds. Metrics include total steps, cadence peaks, minutes and steps in predefined cadence bands, and time and steps in moderate-to-vigorous physical activity (MVPA). Methods and thresholds are informed by the literature, e.g., Tudor-Locke and Rowe (2012) <doi:10.2165/11599170-000000000-00000>, Barreira et al. (2012) <doi:10.1249/MSS.0b013e318254f2a3>, and Tudor-Locke et al. (2018) <doi:10.1136/bjsports-2017-097628>. The package record is also available on Zenodo (2023) <doi:10.5281/zenodo.7858094>.

This package creates and fits staged event tree probability models, which are probabilistic graphical models capable of representing asymmetric conditional independence statements for categorical variables. Includes functions to create, plot and fit staged event trees from data, as well as many efficient structure learning algorithms. References: Carli F, Leonelli M, Riccomagno E, Varando G (2022). <doi: 10.18637/jss.v102.i06>. Collazo R. A., Görgen C. and Smith J. Q. (2018, ISBN:9781498729604). Görgen C., Bigatti A., Riccomagno E. and Smith J. Q. (2018) <arXiv:1705.09457>. Thwaites P. A., Smith, J. Q. (2017) <arXiv:1510.00186>. Barclay L. M., Hutton J. L. and Smith J. Q. (2013) <doi:10.1016/j.ijar.2013.05.006>. Smith J. Q. and Anderson P. E. (2008) <doi:10.1016/j.artint.2007.05.004>.

This package contains tools for survey statistics (especially in educational assessment) for datasets with replication designs (jackknife, bootstrap, replicate weights; see Kolenikov, 2010; Pfefferman & Rao, 2009a, 2009b, <doi:10.1016/S0169-7161(09)70003-3>, <doi:10.1016/S0169-7161(09)70037-9>); Shao, 1996, <doi:10.1080/02331889708802523>). Descriptive statistics, linear and logistic regression, path models for manifest variables with measurement error correction and two-level hierarchical regressions for weighted samples are included. Statistical inference can be conducted for multiply imputed datasets and nested multiply imputed datasets and is in particularly suited for the analysis of plausible values (for details see George, Oberwimmer & Itzlinger-Bruneforth, 2016; Bruneforth, Oberwimmer & Robitzsch, 2016; Robitzsch, Pham & Yanagida, 2016). The package development was supported by BIFIE (Federal Institute for Educational Research, Innovation and Development of the Austrian School System; Salzburg, Austria).

This package provides a set of procedures for parametric and non-parametric modelling of the dependence structure of multivariate extreme-values is provided. The statistical inference is performed with non-parametric estimators, likelihood-based estimators and Bayesian techniques. It adapts the methodologies of Beranger and Padoan (2015) <doi:10.48550/arXiv.1508.05561>, Marcon et al. (2016) <doi:10.1214/16-EJS1162>, Marcon et al. (2017) <doi:10.1002/sta4.145>, Marcon et al. (2017) <doi:10.1016/j.jspi.2016.10.004> and Beranger et al. (2021) <doi:10.1007/s10687-019-00364-0>. This package also allows for the modelling of spatial extremes using flexible max-stable processes. It provides simulation algorithms and fitting procedures relying on the Stephenson-Tawn likelihood as per Beranger at al. (2021) <doi:10.1007/s10687-020-00376-1>.

Finite element modeling (FEM) uses meshes of triangles to define surfaces. A surface within a triangle may be either linear or quadratic. In the order one case each node in the mesh is associated with a basis function and the basis is called the order one finite element basis. In the order two case each edge mid-point is also associated with a basis function. Functions are provided for smoothing, density function estimation point evaluation and plotting results. Two papers illustrating the finite element data analysis are Sangalli, L.M., Ramsay, J.O., Ramsay, T.O. (2013)<http://www.mox.polimi.it/~sangalli> and Bernardi, M.S, Carey, M., Ramsay, J. O., Sangalli, L. (2018)<http://www.mox.polimi.it/~sangalli>. Modelling spatial anisotropy via regression with partial differential regularization Journal of Multivariate Analysis, 167, 15-30.

This package provides a comprehensive framework for building, evaluating, and visualizing regression models for analyzing viral load and CD4 (Cluster of Differentiation 4) lymphocytes data. It leverages the principles of the tidymodels ecosystem of Max Kuhn and Hadley Wickham (2020) <https://www.tidymodels.org> to offer a user-friendly experience in model development. This package includes functions for data preprocessing, feature engineering, model training, tuning, and evaluation, along with visualization tools to enhance the interpretation of model results. It is specifically designed for researchers in biostatistics, computational biology, and HIV research who aim to perform reproducible and rigorous analyses to gain insights into disease dynamics. The main focus is on improving the understanding of the relationships between viral load, CD4 lymphocytes, and other relevant covariates to contribute to HIV research and the visibility of vulnerable seropositive populations.

This package implements an approach aimed at assessing the accuracy and effectiveness of raw scores obtained in scales that contain locally dependent items. The program uses as input the calibration (structural) item estimates obtained from fitting extended unidimensional factor-analytic solutions in which the existing local dependencies are included. Measures of reliability (Omega) and information are proposed at three levels: (a) total score, (b) bivariate-doublet, and (c) item-by-item deletion, and are compared to those that would be obtained if all the items had been locally independent. All the implemented procedures can be obtained from: (a) linear factor-analytic solutions in which the item scores are treated as approximately continuous, and (b) non-linear solutions in which the item scores are treated as ordered-categorical. A detailed guide can be obtained at the following url.

Integrated species distribution modeling is a rising field in quantitative ecology thanks to significant rises in the quantity of data available, increases in computational speed and the proven benefits of using such models. Despite this, the general software to help ecologists construct such models in an easy-to-use framework is lacking. We therefore introduce the R package PointedSDMs': which provides the tools to help ecologists set up integrated models and perform inference on them. There are also functions within the package to help run spatial cross-validation for model selection, as well as generic plotting and predicting functions. An introduction to these methods is discussed in Issac, Jarzyna, Keil, Dambly, Boersch-Supan, Browning, Freeman, Golding, Guillera-Arroita, Henrys, Jarvis, Lahoz-Monfort, Pagel, Pescott, Schmucki, Simmonds and Oâ Hara (2020) <doi:10.1016/j.tree.2019.08.006>.

Computes segregation indices, including the Index of Dissimilarity, as well as the information-theoretic indices developed by Theil (1971) <isbn:978-0471858454>, namely the Mutual Information Index (M) and Theil's Information Index (H). The M, further described by Mora and Ruiz-Castillo (2011) <doi:10.1111/j.1467-9531.2011.01237.x> and Frankel and Volij (2011) <doi:10.1016/j.jet.2010.10.008>, is a measure of segregation that is highly decomposable. The package provides tools to decompose the index by units and groups (local segregation), and by within and between terms. The package also provides a method to decompose differences in segregation as described by Elbers (2021) <doi:10.1177/0049124121986204>. The package includes standard error estimation by bootstrapping, which also corrects for small sample bias. The package also contains functions for visualizing segregation patterns.

Anscombe's quartet are a set of four two-variable datasets that have several common summary statistics but which have very different joint distributions. This becomes apparent when the data are plotted, which illustrates the importance of using graphical displays in Statistics. This package enables the creation of datasets that have identical marginal sample means and sample variances, sample correlation, least squares regression coefficients and coefficient of determination. The user supplies an initial dataset, which is shifted, scaled and rotated in order to achieve target summary statistics. The general shape of the initial dataset is retained. The target statistics can be supplied directly or calculated based on a user-supplied dataset. The datasauRus package <https://cran.r-project.org/package=datasauRus> provides further examples of datasets that have markedly different scatter plots but share many sample summary statistics.

Detection of differentially expressed genes (DEGs) from the comparison of two biological conditions (treated vs. untreated, diseased vs. normal, mutant vs. wild-type) among different levels of gene expression (transcriptome ,translatome, proteome), using several statistical methods: Rank Product, Translational Efficiency, t-test, Limma, ANOTA, DESeq, edgeR. Possibility to plot the results with scatterplots, histograms, MA plots, standard deviation (SD) plots, coefficient of variation (CV) plots. Detection of significantly enriched post-transcriptional regulatory factors (RBPs, miRNAs, etc) and Gene Ontology terms in the lists of DEGs previously identified for the two expression levels. Comparison of GO terms enriched only in one of the levels or in both. Calculation of the semantic similarity score between the lists of enriched GO terms coming from the two expression levels. Visual examination and comparison of the enriched terms with heatmaps, radar plots and barplots.

Compute Buishand Range Test, Pettit Test, SNHT, Student t-test, and Mann-Whitney Rank Test, to identify breakpoints in series. For all functions NA is allowed. Since all of the mention methods identify only one breakpoint in a series, a general function to look for N breakpoint is given. Also, the Yamamoto test for climate jump is available. Alexandersson, H. (1986) <doi:10.1002/joc.3370060607>, Buishand, T. (1982) <doi:10.1016/0022-1694(82)90066-X>, Hurtado, S. I., Zaninelli, P. G., & Agosta, E. A. (2020) <doi:10.1016/j.atmosres.2020.104955>, Mann, H. B., Whitney, D. R. (1947) <doi:10.1214/aoms/1177730491>, Pettitt, A. N. (1979) <doi:10.2307/2346729>, Ruxton, G. D., jul (2006) <doi:10.1093/beheco/ark016>, Yamamoto, R., Iwashima, T., Kazadi, S. N., & Hoshiai, M. (1985) <doi:10.2151/jmsj1965.63.6_1157>.

This package contains all of the functions necessary for the complete analysis of a continuous glucose monitoring study and can be applied to data measured by various existing CGM devices such as FreeStyle Libre', Glutalor', Dexcom and Medtronic CGM'. It reads a series of data files, is able to convert various formats of time stamps, can deal with missing values, calculates both regular statistics and nonlinear statistics, and conducts group comparison. It also displays results in a concise format. Also contains two unique features new to CGM analysis: one is the implementation of strictly standard mean difference and the class of effect size; the other is the development of a new type of plot called antenna plot. It corresponds to Zhang XD'(2018)<doi:10.1093/bioinformatics/btx826>'s article CGManalyzer: an R package for analyzing continuous glucose monitoring studies'.

Calculate Overall Survival or Recurrence-Free Survival for breast cancer patients, using NHS Predict'. The time interval for the estimation can be set up to 15 years, with default at 10. Incremental therapy benefits are estimated for hormone therapy, chemotherapy, trastuzumab, and bisphosphonates. An additional function, suited for SCAN audits, features a more user-friendly version of the code, with fewer inputs, but necessitates the correct standardised inputs. This work is not affiliated with the development of NHS Predict and its underlying statistical model. Details on NHS Predict can be found at: <doi:10.1186/bcr2464>. The web version of NHS Predict': <https://breast.predict.nhs.uk/>. A small dataset of 50 fictional patient observations is provided for the purpose of running examples with the main two functions, and an additional dataset is provided for running example with the dedicated SCAN function.

This package provides estimates for the bivariate and trivariate distribution functions and bivariate and trivariate survival functions for censored gap times. Two approaches, using existing methodologies, are considered: (i) the Lin's estimator, which is based on the extension the Kaplan-Meier estimator of the distribution function for the first event time and the Inverse Probability of Censoring Weights for the second time (Lin DY, Sun W, Ying Z (1999) <doi:10.1093/biomet/86.1.59> and (ii) another estimator based on Kaplan-Meier weights (Una-Alvarez J, Meira-Machado L (2008) <https://w3.math.uminho.pt/~lmachado/Biometria_conference.pdf>). The proposed methods are the landmark estimators based on subsampling approach, and the estimator based on weighted cumulative hazard estimator. The package also provides nonparametric estimator conditional to a given continuous covariate. All these methods have been submitted to be published.

Finds single- and two-arm designs using stochastic curtailment, as described by Law et al. (2022) <doi:10.1080/10543406.2021.2009498> and Law et al. (2021) <doi:10.1002/pst.2067> respectively. Designs can be single-stage or multi-stage. Non-stochastic curtailment is possible as a special case. Desired error-rates, maximum sample size and lower and upper anticipated response rates are inputted and suitable designs are returned with operating characteristics. Stopping boundaries and visualisations are also available. The package can find designs using other approaches, for example designs by Simon (1989) <doi:10.1016/0197-2456(89)90015-9> and Mander and Thompson (2010) <doi:10.1016/j.cct.2010.07.008>. Other features: compare and visualise designs using a weighted sum of expected sample sizes under the null and alternative hypotheses and maximum sample size; visualise any binary outcome design.

Facilitates creation and manipulation of metric graphs, such as street or river networks. Further facilitates operations and visualizations of data on metric graphs, and the creation of a large class of random fields and stochastic partial differential equations on such spaces. These random fields can be used for simulation, prediction and inference. In particular, linear mixed effects models including random field components can be fitted to data based on computationally efficient sparse matrix representations. Interfaces to the R packages INLA and inlabru are also provided, which facilitate working with Bayesian statistical models on metric graphs. The main references for the methods are Bolin, Simas and Wallin (2024) <doi:10.3150/23-BEJ1647>, Bolin, Kovacs, Kumar and Simas (2023) <doi:10.1090/mcom/3929> and Bolin, Simas and Wallin (2023) <doi:10.48550/arXiv.2304.03190> and <doi:10.48550/arXiv.2304.10372>.