mrgsolve blog – mrgsolve/blog

Simulate true Cmax with ODE

This post outlines some considerations when simulating a true Cmax in your PK model using differential equations.

Cmax

ODE

It’s a common practice to find a true maximum concentration after a dose (Cmax) through simulation from a PK model written using differential equations. In this post, we’ll…

Patterns for coding doses inside a model

This is a vignette showing worked examples implementing modeled doses or other events with the evtools plugin. All code is provided in a GitHub repository linked in the document.

evtools

dynamic-dosing

Dynamic dosing in pharmacokinetic (PK) or pharmacokinetic pharmacodynamic (PKPD) models allows for changes in the dose regimen in response to changes in model state…

New in version 1.5.1

mwrite

evtools

dynamic-dosing

pkmodel

data set

new release

mrgsolve version 1.5.1 was released to CRAN in July, 2024. This release features new functions to write a model back to a file in either yaml format or native mrgsolve.

Estimate a multi-state model with mrgsolve

evtools

msm

example

draft

This post shows how you can fit a simple multi-state model with mrgsolve. We’ll fit a model to the cav data from the msm package using both msm and mrgsolve and then compare…

Dynamic dosing with mrgsolve and evtools

dynamic-dosing

evtools

The evtools plugin is a set of functions and classes you can use to implement dosing regimens from inside your model. It first became available in mrgsolve 1.4.1.

Update: Validate translation from NONMEM

validation

With mrgsolve version 1.0.8, there is a more complete way to validate your model against a completed NONMEM estimation run.

New in 1.0.8

new release

Find out what’s new in mrgsolve version 1.0.8

Import estimates from multiple NONMEM runs

You can use $NMEXT and $NMXML to import estimates from multiple NONMEM runs into a single mrgsolve model.

Validate model translation from NONMEM

model specification

validation

If your mrgsolve model is a translation of a completed NONMEM model run, you can validate it every time; find out how.

Stopping long-running simulations

stop

Sometimes, you want to stop a simulation that is running (or might run) longer than you want. Find out how.

New in 1.0.0

new release

mrgsolve celebrates her version 1.0.0 release. Find out about some very cool and powerful features that we included in this release.

Simulate From EBE

EBE

This blog post will show you how to use a single model to do both population simulation and simulation from a specific set of empirical Bayes estimates. The basic implementation for this approach was pioneered by Tim Waterhouse at MetrumRG.

New in 0.11.0 and 0.11.1

new release

Find out what’s new in mrgsolve version 0.11.0 and 0.11.1.

New in 0.10.7

new release

Find out what’s new in mrgsolve version 0.10.7.

New in 0.10.0 and 0.10.1

new release

Find out what’s new in mrgsolve version 0.10.0 and 0.10.1.

Zero-order steady state infusion

events

infusion

steady state

Starting with version 0.10.0, mrgsolve can implement dosing records calling for a zero-order infusion at steady state. Find out more in this blog post.

Steady State in mrgsolve

steady state

This blog post describes the mechanism by which mrgsolve finds steady state under this new change set (starting late 2019 to early 2020) and describes some of the settings…

Time after dose (TAD)

time after dose

tad

In this post, we’ll look at different ways you can bring time after dose into your simulated outputs.

Alternate dosing event syntax

events

A fun syntax for creating event objects for different dosing regimens.

Validating Simulation Results with pmxTools

validation

In this blog post, we run some simulations with mrgsolve and the excellent pmxTools package and compare results.

Rmarkdown model specification

model specification

I came up with this fun model specification format that looks just like Rmarkdown code chunks.

Tidy workflow for simulating dynamics in R

example

I worked an example for a SIR model to show people in the ecology field how they can use mrgsolve for their ODE-based models.

Modeled interventions in mrgsolve

modeled events

mtime

This post is to introduce modeled interventions in mrgsolve. The main use case is to force mrgsolve to advance the system to a specific time so that some aspect of the system can change at that time.

mtime

modeled events

mtime

This is just a post on a simplified implementation of modeled events in mrgsolve. There is also another, more complete blog post on this topic published previously .

Models without compartments

pred

model specification

Introducing a new model block ($PRED) that lets you write models with analytical solutions.

Generating input data sets for mrgsolve

data set

An important mechanism for creating robust, complex simulations is the input data set. In this blog post, we’ll look at some ways you can quickly create input data sets for your simulation.

Mixture model example

mixture model

In this post, we’ll look at how you can code up a mixture model in mrgsolve.

A Complete Example

example

We had a user who was learning mrgsolve ask for a “complete example”. I wasn’t sure what exactly that meant, but I created this example and I’m sharing it today on the blog.

Reproducible results with set.seed

example

This post demonstrates that random simulation outputs can be made reproducible by using set.seed().

Generate MAP Bayes Parameter Estimates

MAP Bayes

This tutorial illustrates now to do MAP Bayes estimation with mrgsolve.

MAP Bayes example

MAP Bayes

optimization

This is a shortened version of map_bayes.html, showing only the estimation step.

Work with event objects

events

event objects are data.frame-like data structures that allow you to quickly and flexibly include dosing interventions into your simulations.

Set initial conditions

initial conditions

There are several ways to set initial conditions in mrgsolve. Find out how this can work to your advantage when planning your simulations.