An open-source pharmacometrician’s workflow in R: from exploration (xGx) to model building (nlmixr) and diagnostics (ggPMX)

PAGE 2022

By Rik Schoemaker in workshop

June 22, 2022

Workshop target audience

Pharmacometricians/modelers with basic knowledge on model building, evaluation and qualification. Basic knowledge of writing and executing R scripts is required.

Workshop overview

The workshop will provide a tutorial on three open–source R packages, supporting the pharmacometrics workflow in exploring and modeling clinical data:

  • Exploration of the data using the Exploratory Graphics (xgxr) package, available on CRAN and GitHub, and here.
  • Population PK and PKPD modeling of the data using nlmixr2 package, available on CRAN and GitHub, and here. nlmixr2 builds on the ODE simulation package rxode2, by implementing parameter estimation algorithms like SAEM and FOCE with interaction.
  • Model building and validation using ggPMX, a library of reproducible diagnostic plots available on CRAN and on GitHub.

The combination of the three open-source R packages provides the pharmacometrics modeling community the opportunity to reduce the learning curve needed to become proficient on each of the different tasks using a stepwise framework.

Workshop Learning Objectives

During the workshop, the participants will have the opportunity to become familiar with the packages with extensive hands-on sessions, which will follow the initial lectures on xgxr, ggPMX, and nlmixr2. The participants will have a chance to experience the stepwise framework of the Pharmacometrics workflow. First, through a question-based approach, xgxr helps to uncover useful insights that can be revealed without complex modelling and to identify aspects of the data that may be explored further. Next, nlmixr2 is used for building an adequate population model refined by the exploration of the data to characterize the dose-exposure-response relationship. Finally, the model evaluation, validation and reporting is driven by ggPMX. The model diagnostic plots help in selecting the model describing the data most accurately.

We’re going to be delivering this course on Tuesday 28 June in Ljubljana! We’ll expand this page with photos and suchlike after the course has been given.

Partner materials

Our colleagues in the ggPMX team have made their course materials for this session available here!

Posted on:
June 22, 2022
Length:
2 minute read, 322 words
Categories:
workshop
See Also: