Info

Source code for the paper:

Siivola E, Weber S, Vehtari A. Qualifying drug dosing regimens in pediatrics using Gaussian processes. Statistics in Medicine. 2021;1–18. https://doi.org/10.1002/sim.8907

Link to the online version of the paper https://onlinelibrary.wiley.com/doi/full/10.1002/sim.8907

NOTE: Stan >= 2.11 is required

Quick Start

Run oral_1cmt_allo.R in src folder

Introduction

These files are an attempt to model maturation in clearance with the help of Gaussian Processes. The code for PKPD is mostly based on Sebastian Webers' work on building PKPD models in Stan. Pharmacokinetic (PK) models describe the relationship of drug concentration in a given subject over time as a function of the dosing history. PK models facilitate mass action kinetics laws to describe the drug absorption, distribution and elimination process with a compartmental approximation.

NONMEM Data Sets

A widely known data format to describe the longitudinal data of each patient is the NONMEM format. As such convenient functions are provided to map a NONMEM data set to the internal format.

Files

• src/stan/*.stan different stan model definitions
• src/stan/utils/* helper functions for Stan
• src/utils.R functions for generating data, setting defaults and plotting
• src/oral_1ct_allo.R the main file