On behalf of our entire group, we are excited to announce that Pmetrics 3 is now live! While there is not much change with respect to the R framework compared to Pmetrics 2, we have done so much under the hood in our transition to modern, efficient, fast code.

Simplified model fitting

The biggest change in R is that we have simplified model fitting to only two steps:

1. Create a model, which automatically compiles it.
2. Use the model’s “fit” method to estimate its parameter value distributions from the data, i.e., “run” the model.

The simplified workflow is detailed in our entirely new online Pmetrics book.

Other new features

• Handling observations that are below or above limits of quantification
• Independent error models (gamma/lambda) for each output equation
• Eliminating constraints on population size, number of random parameters, and number of support points in discrete simulation priors
• Allowing additional doses to be added before or after time 0
• Converting plotly interactive plots to static ggplots with plotlygg(), our custom complementary function to plotly::ggplotly().

Coming soon

Our Pmetrics graphical user interface, a beautiful tool for beginners and experienced users alike that will complement the R package.

Updated Documentation

Complete, updated documentation is now available at https://lapkb.github.io/Pmetrics/, including:

• Installation
• Getting started
• Function reference

Open Source on GitHub

Full open source code, issue tracking, and discussion boards are available on GitHub:

• View and contribute to the source code
• Report bugs and request features
• Join community discussions
• Access development history

Citation

If you use Pmetrics in your research, please cite:

Neely MN, van Guilder MG, Yamada WM, Schumitzky A, Jelliffe RW. Accurate Detection of Outliers and Subpopulations With Pmetrics, a Nonparametric and Parametric Pharmacometric Modeling and Simulation Package for R. Therapeutic Drug Monitoring. 2012; 34(4): 467-476.