Registration for the 2026 workshop “R for pharmacometricians” opened!
Dear Members,
The GMP is thrilled to invite you to our upcoming online 2026 R for pharmacometricians workshop !
R is a versatile language and environment used for statistical computing and graphics. It offers a wide range of statistical techniques such as linear and nonlinear modeling, classical statistical tests, time-series analysis, classification, clustering, and more. Additionally, R’s graphics capabilities allow the production of well-designed, publication-quality plots, complete with mathematical symbols and formulas.
An important advantage of R is its versatility. It provides an open-source route for participating in statistical methodology research, with the S language being a popular choice in this field.
R can be freely accessed as Free Software under the GNU General Public License. It can be compiled and run on various UNIX platforms, Windows, and MacOS.
The objective of this workshop is to provide beginners with solid foundations in data management using R and the tidyverse, and to introduce more advanced participants to machine learning and the development of Shiny applications.
Registration
The registration to this workshop is opened to Academic, Industrial, Regulatory, and Student members and will give access to the 3 sessions. To become a member, go to the member area. To register, purchase your seat here.
Agenda
This workshop will be split in 3 remote sessions as follows:
– Session 1: Data Management with R and the Tidyverse (June, 11th 2026, 2:00 p.m. CET) by Alexandre Destere (CHU Nice) – 3 hours including breaks
– Session 2: Introduction to Machine Learning with R and tidymodels (June, 24th 2026, 2 p.m. CET) by Jean-Baptise Woillard (université de Limoges) – 3 hours including breaks
– Session 3: Rapid PK/PD Simulations & Interactive Apps with Campsis and Shiny (September, 15th 2026, 2 p.m. CET) by Nicolas Luyckx (Calvagone) – 3 hours including breaks
Details about the content of each session
Session 1 – Data Management with R and the Tidyverse
Overview
Data is rarely clean. Before any meaningful analysis can take place, raw datasets must be imported, inspected, reshaped, and validated — and doing this in a reproducible, readable way is a skill in itself. This workshop is designed for scientists who want to move beyond manual spreadsheet editing and build reliable data pipelines in R.
Using the tidyverse — a coherent collection of R packages sharing a common design philosophy — we will cover the full data management workflow from raw file import to a clean, analysis-ready table. The focus is not on memorizing functions, but on understanding the logic of a tidy pipeline so you can apply it immediately to your own datasets.
What You Will Learn
By the end of this session, you will be able to:
- Import: Read CSV, Excel, and text files into R and handle common issues such as encoding, missing headers, and mixed data types.
- Inspect: Quickly explore the structure and quality of a dataset using tidyverse-native tools.
- Clean: Filter invalid rows, rename variables, recode categorical values, and handle missing data with dplyr.
- Reshape: Convert between wide and long formats using tidyr to suit any downstream analytical need.
- Summarize: Compute grouped descriptive statistics and build reproducible summary tables.
- Join: Merge data from multiple sources and detect mismatches using dplyr join functions.
- Export: Save your cleaned dataset and organize your project following reproducibility best practices.
Target Audience
This workshop is ideal for students and researchers who have some basic familiarity with R but have not yet worked with the tidyverse. No prior experience with the packages is required, and no advanced statistical background is needed — the session focuses entirely on data handling, not modelling. It is also well suited for anyone currently relying on Excel for data preparation and looking for a more scalable and reproducible alternative.
Prerequisites
No prior experience with the tidyverse is expected. Participants should feel comfortable with the very basics of R — knowing what a vector and a data frame are, and having run at least a few R scripts before is sufficient. A working installation of R (version 4.2 or later) and RStudio or Positron is required. All packages (Tidyverse) will be installed together at the start of the session.
Overview
Machine learning offers powerful tools for building predictive models, but moving from a raw dataset to a robust and interpretable workflow can be challenging. This workshop is designed for scientists who want a practical introduction to machine learning in R, with a focus on building end-to-end modeling pipelines using the tidymodels ecosystem.
What You Will Learn
By the end of this session, you will be able to:
- Prepare a dataset for machine learning in R, including reshaping, merging outcomes, filtering, and creating informative predictors.
- Explore the relationships between predictors and outcomes using graphical tools.
- Build preprocessing pipelines with recipes, including variable selection and normalization steps.
- Train several predictive models within the tidymodels framework using a consistent workflow structure.
- Tune model hyperparameters using cross-validation and compare model performance with appropriate metrics such as RMSE.
- Evaluate and finalize a model that can be used for prediction on new data.
Target Audience
This session is intended for researchers, pharmacometricians, and data analysts who have a good knowledge of R and tidyverse ecosystem (or have followed Session 1) and want to get started with applied machine learning in a structured and reproducible way. No prior experience with tidymodels is required. The session will be especially useful for participants who want to understand how to organize a complete modeling workflow rather than focus on a single algorithm in isolation.
Prerequisites
Having a good knowledge of R and tidyverse ecosystem (or followed Session 1) is expected.
Session 3 – Rapid PK/PD Simulations & Interactive Apps with Campsis and Shiny
Overview
Simulation is at the heart of modern pharmacometrics, but the technical overhead can often be a barrier. This workshop is designed for scientists who want to move quickly from a model idea to an interactive simulation tool.
Using Campsis, a powerful and user-friendly R package for PK/PD simulation, we will bridge the gap between coding a model and visualizing its outcomes. We will then take it a step further by integrating these simulations into Shiny to create dynamic, user-facing applications.
The focus is not on mastering every technical detail of the package, but on learning a streamlined workflow that you can apply to your own projects—regardless of which simulation engine you eventually choose to use.
What You Will Learn
By the end of this session, you will be able to:
- Implement: Build a PK/PD model from scratch within the Campsis environment.
- Design: Create simulation protocols (dosing regimens, covariates) using Campsis’ intuitive scripting language.
- Execute & Analyze: Run simulations and efficiently calculate key PK metrics and visual summaries.
- Deploy: Transition your R code into a functional Shiny application.
- Interact: Use Shiny components (sliders, inputs) to dynamically modify model parameters or designs and see results in real-time.
Target Audience
This workshop is ideal for pharmacometricians and researchers who have a basic grasp of R and want to accelerate their simulation workflow.
Prerequisites
No prior experience with Campsis is required—the package is designed to be accessible, allowing us to focus on the “how-to” rather than complex syntax.
Before attending the webinar series, please be sure to install R and RStudio on your computer (seek help from your IT department if needed). Installation guides for R and RStudio are available here and here, respectively.