PKanalix documentation

Version 2020

This documentation is for PKanalix starting from 2019 version.
PKanalix performs analysis on PK data set including:

  • The non compartmental analysis (NCA), which computes NCA parameters using the calculation of the \(\lambda_z\) – slope of the terminal elimination phase.
  • The compartmental analysis (CA), which finds parameters of a model representing the PK as the dynamics in compartments for each individual. It does not include population analysis that could be performed in Monolix.

What else?

  • A clear user-interface with a simple workflow to efficiently run the NCA and CA analysis.
  • Easily accessible PK models library and auto-initialization method to improve the convergence of the optimization of CA parameters.
  • Automatically generated results and plots to give an immediate feedback.
  • Interconnection with MonolixSuite application to export of project to Monolix for the population analysis.

and, starting from the 2020 version:

  • Filters of a dataset to easily perform the analysis on several data subsets without modifying the original file.
  • Module to select and scale output units for better analysis and reporting.
  • Flexibility: selection of NCA parameters for computation and display, stratification of results by categorical covariates and acceptance criteria for comparison, multiple partial AUC
  • Plots with statistics of observed data.
  • More interface features: dark theme, font size, choice of significant digits.

PKanalix tasks

Pkanalix uses the dataset format common for all MonolixSuite applications, see here for more details. It allows to move your project between applications, for instance export a CA project to Monolix and perform a population analysis with one “click” of a button. Based on a dataset, there are two tasks:

Non Compartmental analysis

The first main feature of PKanalix is the calculation of the parameters in the Non Compartmental Analysis framework.

This task consists in defining rules for the calculation of the \(\lambda_z\) (slope of the terminal elimination phase) to compute all the NCA parameters. This definition can be either global via general rules or manual on each individual, where the user selects or removes  points in the \(\lambda_z\) calculation.

Compartmental Analysis

The second main feature of PKanalix is the calculation of the parameters in the Compartmental Analysis framework. It consists in finding parameters of a model that represents the PK as the dynamics in compartments for each individual.

This task defines a structural model (based on a user-friendly PK models library) and estimates the parameters for all the individuals. Automatic initialization method improves the convergence of parameters for each individual.

All the NCA and/or CA outputs are automatically displayed in sortable tables in the Results tab. Moreover, they are exported in the result folder in a R-compatible format. Interactive plots give an immediate feedback and help to better interpret the results.

The usage of PKanalix is available not only via the user interface, but also via R with a dedicated R-package (detailed here). All the actions performed in the interface have their equivalent R-functions. It is particularly convenient for reproducibility purpose or batch jobs.