The Expo demonstrates using bbr.bayes and NONMEM® in a typical Bayesian (Bayes) population pharmacokinetic (pop PK) modeling and simulation (M&S) analysis, including model fitting, model evaluation, and model summarization. This demonstration uses the same processes and suite of tools used at Metrum Research Group (MetrumRG) to ensure traceable and reproducible pharmacometrics research; however, it is not meant to be a complete vignette on using all of the features of bbr.bayes or the other tools used in the workflow. Links are provided in the tools section of the expo that shares additional information about each of the tools used.
What you’ll find in this Expo:
A description of the data and statistical models fitted to the data.
Our approach to M&S activities and reporting.
Access to example code in a GitHub repository.
Information and vignettes on MetrumRG’s suite of tools.
This Expo is not intended to be a comprehensive tutorial on the theory and application of Bayes analyses in NONMEM®. For such a tutorial, we refer readers to the following publication:
Johnston CK, Waterhouse T, Wiens M, Mondick J, French J, Gillespie WR. Bayesian estimation in NONMEM. CPT Pharmacometrics Syst Pharmacol. 2023; 00: 1-16. doi.org/10.1002/psp4.13088
The diagram below illustrates the end-to-end process of a typical modeling process from setup through completion. (Click through to visit the PDF with hyperlinks.)
The general workflow in this Expo was conducted primarily in the R programming language and used NONMEM® to fit the models. This analysis was performed on Metworx, our platform as a service (PaaS), for high-performance elastic cloud computing; however, Metworx is not required to use the tools and processes illustrated here.
For comments, questions, or more information about any of the tools or work process demonstrated here, please contact us.