This list covers the wider purpose R packages that we have found most useful and those that most commonly form dependencies to the pharma-specific packages.
Another useful reference is this document from the R Foundation titled “Regulatory Compliance and Validation Issues - A Guidance Document for the Use of R in Regulated Clinical Trial Environments”, which shows a list of base and recommended R packages.
There is a current industry working group ongoing effort around tables and the advantages of different packages here, which will result in a paper. In general, the problem is best divided into the following categories:
Creates tables for HTML, PDF, Microsoft Word and PowerPoint documents from R Markdowngt
Easily generate information-rich, publication-quality tables from Rhuxtable
An R package to create styled tables in multiple output formats, with a friendly, modern interface.mmtable2
Create and combine tables with a ggplot2/patchwork syntax.
Formatting and Rendering
There are different solutions already available, but during R/Pharma there was an announcement of the RStudio package tgen to solve the rendering problems consistently.
The goal of visR is to enable fit-for-purpose, reusable clinical and medical research focused visualizations and tables with sensible defaults and based on sound graphical principles.ggplot2
An implementation of the Grammar of Graphics in R, and the most popular plotting package for static plots in R.viridisLite
Colorblind-Friendly Color Maps for R
Set of packages with a common API for data manipulation and TLGs (includes dplyr, ggplot2, etc)diffdf
The diffdf package is designed to enable detailed comparison of two data.frames.lintr
Static code analysis for R that checks adherence to a given style, syntax errors and possible semantic issuesrenv
Helps you create reproducible environments for your R projects
It would be impossible to capture all the relevant statistical packages, as there are potentially thousands of potential methods that could be applied in a study. Below we include a handful that are likely to be used in the majority of trial reporting events, and refer to the Cran Task Views (CTVs) that attempt to maintain a landscape of available packages other several important domains.
stats - Base R package with a lot of functionality useful for design and analysis of clinical trials
survival - Core survival analysis routines
car - R companion to applied regression
emmeans - Estimated marginal means, aka least-squares means
mmrm - Mixed models for repeated measures (link)
lme4 - Fit linear and generalized linear mixed-effects models
lmerTest - Tests in linear mixed effects models
broom - takes the messy output of built-in functions in R and turns them into tidy tibbles
VGAM - Vector generalized linear and additive models
It is also worth attracting attention here to the wider CTVs across several pertinent topics, such as:
As you can appreciate with R such an ever-evolving language we would never be able to include and maintain a complete list of all recommended packages here, but we hope that this page helps to introduce to some of the most common packages used both within and beyond pharma.