FeRx NLME - R package (ferx)

ferx is a fast, ergonomic R package for nonlinear mixed effects (NLME) modeling, powered by a Rust backend (ferx-core). It targets the pharmacokineticist who wants NONMEM-like expressiveness with modern tooling: a plain-text model DSL, tidy R output, and a solver fast enough to iterate quickly.

What this book covers

This book walks through ferx from first install to advanced modeling techniques. It is organized into three parts:

Getting started covers installation and a complete quickstart example.

Core workflow goes deep on the model DSL, fitting options, diagnostic output, and simulation/VPC — the four things you do on every project.

Advanced topics covers parameter transforms, ODE models, BLOQ/M3 censoring, and inter-occasion variability.

Every code chunk in this book is executed on CI and the output shown is real. Where fits are slow, results are cached with cache: true; you can always clear the cache and re-run.

Prerequisites

  • R ≥ 4.2
  • Rust toolchain (see Prerequisites)
  • Basic familiarity with population PK/PD modeling (one-compartment models, ETAs, RUV)

Companion resources