What is PAGE?

We represent a community with a shared interest in data analysis using the population approach.

is a general-purpose program for Bayesian statistical analysis. BUGS version 0.6 can handle a wide range of models including linear mixed-effect models. The 1.4 version of WinBUGS is currently available and is capable of fitting non-linear mixed effect models; a specialist Windows interface for PK modelling (PKBUGS) is included.
is a program that implements the SAEM algorithm for estimation in nonlinear mixed-effects models with several nice new features: library of usual PK and PD models as well as discrete data models, additional graphical output, enhanced interface, handling of measurements below limit of quantification, MLXTRAN language for easy model description, etc. Available as a MATLAB program or as a stand-alone executable version that does not require MATLAB.
is a freely available open-source package for R that does not depend on any commercial software, and is available on CRAN and GitHub. The package allows structural models to be implemented using a system of ordinary differential equations (ODEs), and allows fully flexible dosing definitions in terms of the type (e.g. bolus doses or infusions), the timing, the number of doses, and their amount, which can vary between individuals. nlmixr builds on RxODE, a fast and efficient R package for simulating nonlinear mixed effect models using ODEs, with rapid execution due to compilation in C. Comprehensive online documentation is available, along with an nlmixr tutorial in CPT:PSP, and a performance comparison paper in CPT:PSP. The package comes with its dedicated project manager shinyMixR that runs in a web-browser, and is linked to xpose for graphical exploration and goodness of fit plots.
is a powerful and well-established tool for analysing population PK data as well as fitting general linear and nonlinear models. Originally developed by Lewis B. Sheiner and Stuart Beal at the University of California at San Francisco, it is now commercially available from ICON Clinical Research LLC. Some sueful resources -
Phoenix NLME
is Certara's powerful data processing and modeling software for population PK/PD analysis. Phoenix NLME includes flexible tools to create models from an extensive built-in library of models, graphically, or via user-supplied code. Features include graphical workflows that integrate data preparation, model selection and generation of tables and figures, and parallelized model estimation engines designed to take advantage of significant speed gains available on modern computers running multiple processors. Phoenix NLME can be optionally configured to connect with a drug development organization's computer grids and clusters.
is a suite of Population PKPD tools, implemented in the popular and high-level Python language, that is both powerful enough to use for complex nonlinear mixed-effects modelling and simple enough to use as a teaching and learning tool. PoPy fits a parametric model to population PK and PD data via a range of minimization methods (including Quasi-Newton, Expectation-Maximization, and novel algorithms that avoid derivatives), likelihood distributions (including normals, truncated normals for BLQ observations, and discrete distributions such as Bernoulli), dosing functions (including Weibull and Gamma) and solvers (both closed-form solutions and Ordinary Differential Equations). It exploits distributed processing, will simulate novel datasets over multiple populations in order to analyse the recovery of ground truth parameters, and presents all results in an easy-to-read format. PoPy benefits from the huge universe of Python libraries that enable active and rapid development, uses an intuitive syntax that makes it easy to express complex hierarchical models (e.g., with inter-occasion variability), comes with extensive documentation and tutorials, and is free for non-commercial and academic use.
is a one-stop integrated modelling and simulation platform powered by the Julia language. It can perform non-compartmental analysis (NCA), nonlinear mixed-effects modelling (NLME), Bioequivalence (BE), in vitro-in vivo correlation (IVIVC) and clinical trial simulations (CTS). Further, Pumas provides convenient way of handling of multi-scale PBPK, QSP models. DeepNLME is the state-of-the-art product of Pumas that seamlessly integrates NLME and Deep Learning, via scientific machine learning. Pumas comes with a full suite of productivity tools to pre- and post-process your analyses in an interactive manner. Scientists can easily convert complex models to dashboards effortlessly to collaborate with interdisciplinary teams. Pumas was developed as a cloud-first technology that allows scientists scale to thousands of CPUs and GPUs with a single click. The software is free for non-commercial research and training.
is a free and open-source probabilistic programming language implementing full Bayesian statistical inference. See mc-stan.org for interfaces to R, Matlab, Python, cmdline, Julia, Stata and Mathematica, find the reference manual and case studies at mc-stan.org/documentation/, and get help from the Stan users mailing list.

Please send comments, requests for changes or additions and other questions related to this list to Justin Wilkins