Type:	Package
Title:	Application of NODEs in 'Monolix', 'NONMEM', and 'nlmixr2'
Version:	0.1.0
Description:	An easy-to-use tool for implementing Neural Ordinary Differential Equations (NODEs) in pharmacometric software such as 'Monolix', 'NONMEM', and 'nlmixr2', see Bräm et al. (2024) <doi:10.1007/s10928-023-09886-4> and Bräm et al. (2025) <doi:10.1002/psp4.13265>. The main functionality is to automatically generate structural model code describing computations within a neural network. Additionally, parameters and software settings can be initialized automatically. For using these additional functionalities with 'Monolix', 'pmxNODE' interfaces with 'MonolixSuite' via the 'lixoftConnectors' package. The 'lixoftConnectors' package is distributed with 'MonolixSuite' (https://monolixsuite.slp-software.com/r-functions/2024R1/package-lixoftconnectors) and is not available from public repositories.
License:	GPL (≥ 3)
Encoding:	UTF-8
RoxygenNote:	7.3.3
Suggests:	rxode2, nlmixr2, lixoftConnectors, withr, testthat (≥ 3.0.0), knitr, rmarkdown
Imports:	tidyr, ggplot2, checkmate
Config/testthat/edition:	3
BugReports:	https://github.com/braemd/pmxNODE/issues
VignetteBuilder:	knitr
NeedsCompilation:	no
Packaged:	2025-11-19 13:00:17 UTC; BraemD
Author:	Dominic Bräm [aut, cre], Bernhard Steiert [ctb], Gilbert Koch [ctb], Matthew Fidler [ctb]
Maintainer:	Dominic Bräm <domi.braem@hotmail.com>
Repository:	CRAN
Date/Publication:	2025-11-26 09:20:02 UTC

Neural Network ODE language in nlmixr2 langauage

Description

Neural Network ODE language in nlmixr2 langauage

Usage

NN(
  number = 1,
  state = "t",
  min_init = 0.5,
  max_init = 10,
  n_hidden = 5,
  act = c("ReLU", "Softplus"),
  time_nn = FALSE,
  beta = 20,
  pop = getOption("pmxNODE.pop", TRUE),
  eta_model = c("prop", "add"),
  theta_scale = 0.1,
  eta_scale = 0.1,
  pre_fixef = getOption("pmxNODE.pre_fixef", NULL),
  iniDf = NULL
)

Arguments

Value

A list with the before and replace elements and iniDf to allow integration in the rxode2/nlmixr2 language directly.

Author(s)

Matthew L Fidler (uses the same functions 'nn_generator_nlmixr', written by Dominic Bräm)

Examples

## Not run: 
 if (requireNamespace("rxode2", quietly = TRUE)) {

# Called directly, this isn't that interesting, but can show what
# is produced for rxode2 integration

library(rxode2)

NN(1, state="t", min_init=0.1, max_init=24, pop=TRUE)

# This can be used in the rxode2 language as follows:

f_ode_pop <- function(){
  ini({
    lV <- 1
    prop.err <- 0.1
  })
  model({
    V <- lV
    d/dt(centr)  =  NN(1, state=centr,min_init=0,max_init=300)
    cp = centr / V
    cp ~ prop(prop.err)
  })
}

# but it expands to the complete model:

f_ode_pop()

# This is because pmxNODE uses the extensible user model interface
# in rxode2.  This only works if you load rxode2/nlmixr2 and pmxNODE

}

## End(Not run)

Change a population Neural Network model to a model with between subject variability

Description

This only changes the Neural Network model to add between subject variability. It assumes the following parameter structure

Usage

NNbsv(ui, val = 0.1, str = "%s <- l%s*exp(eta.%s)", warn = FALSE)

Arguments

Value

modified model with between subject variabilities added for neural-network components.

Author(s)

Matthew L. Fidler

Examples

## Not run: 
f_ode_pop <- function(){
  ini({
    lV <- 1
    prop.err <- 0.1
  })
  model({
    V <- lV
    d/dt(centr)  =  NN(1, state=centr,min_init=0,max_init=300)
    cp = centr / V
    cp ~ prop(prop.err)
  })
}

f_ode_pop() %>% NNbsv(.2, warn=TRUE)

## End(Not run)

Copy examples to your folder

Description

This function allows to copy one or multiple examples (data and model files) to a directory of your choice.
Either examples, example_nr, example_software, or example_nr + example_software must be given.

Usage

copy_examples(
  target_folder,
  examples = NULL,
  example_nr = NULL,
  example_software = NULL,
  pkg_name = "pmxNODE"
)

Arguments

Value

Copied examples in specified folder.

Author(s)

Dominic Bräm

Examples

## Not run: 
copy_examples("path/to/target/folder",
              examples = c("data_example1_mlx.csv","mlx_example1_model.txt"))
copy_examples("path/to/target/folder",example_nr = 1)
copy_examples("path/to/target/folder",example_software = "Monolix")
copy_examples("path/to/target/folder",example_nr = 1, example_software = "NONMEM")

## End(Not run)

Generate Derivative versus State Plot (Monolix)

Description

This functions allows to generate a derivative versus state plot for a neural network from a NODE in Monolix.

Usage

der_state_plot_mlx(
  nn_name,
  min_state = NULL,
  max_state = NULL,
  inputs = NULL,
  est_parms = NULL,
  mlx_file = NULL,
  time_nn = FALSE,
  act = "ReLU",
  length_out = 100,
  plot_type = c("base", "ggplot"),
  beta = 20,
  transform = NULL
)

Arguments

Details

Either est_parms or mlx_file must be given. If both arguments are given, est_parms is prioritized.

Value

Displaying derivative versus state plot; returns ggplot-object if plot_type="ggplot"

Author(s)

Dominic Bräm

Examples

mlx_path <- system.file("extdata","mlx_example1_ind.mlxtran",package="pmxNODE")
der_state_plot <- der_state_plot_mlx(nn="c",
                                     min_state=0,max_state=10,
                                     mlx_file=mlx_path,
                                     plot_type="ggplot")

Generate Derivative versus State Plot (nlmixr2)

Description

This functions allows to generate a derivative versus state plot for a neural network from a NODE in nlmixr2

Usage

der_state_plot_nlmixr(
  nn_name,
  min_state = NULL,
  max_state = NULL,
  inputs = NULL,
  est_parms = NULL,
  fit_obj = NULL,
  length_out = 100,
  time_nn = FALSE,
  act = "ReLU",
  plot_type = c("base", "ggplot"),
  beta = 20
)

Arguments

Details

Either est_parms or fit_obj must be given. If both arguments are given, est_parms is prioritized.

Value

Displaying derivative versus state plot; returns ggplot-object if plot_type="ggplot"

Author(s)

Dominic Bräm

Examples

## Not run: 
pop_fit <- nlmixr2(node_model_pop,data=data,est="bobyqa")
der_state_plot <- der_state_plot_nlmixr(nn="c",
                                        min_state=0,max_state=10,
                                        fit_obj=pop_fit,
                                        plot_type="ggplot")

## End(Not run)

Generate Derivative versus State Plot (NONMEM)

Description

This functions allows to generate a derivative versus state plot for a neural network from a NODE in NONMEM. Can also be used for nlmixr2.

Usage

der_state_plot_nm(
  nn_name,
  min_state = NULL,
  max_state = NULL,
  inputs = NULL,
  est_parms = NULL,
  nm_res_file = NULL,
  length_out = 100,
  time_nn = FALSE,
  act = "ReLU",
  plot_type = c("base", "ggplot"),
  beta = 20,
  transform = NULL
)

Arguments

Details

Either est_parms or nm_res_file must be given. If both arguments are given, est_parms is prioritized.

Value

Displaying derivative versus state plot; returns ggplot-object if plot_type="ggplot"

Author(s)

Dominic Bräm

Examples

res_path <- system.file("extdata","nm_example1_model_converted_ind.res",package="pmxNODE")
der_state_plot <- der_state_plot_nm(nn="c",
                                    min_state=0,max_state=10,
                                    nm_res_file=res_path,
                                    plot_type="ggplot")

Generate Derivative versus State (Monolix)

Description

This functions allows to generate derivative versus state data for a neural network from a NODE in Monolix.

Usage

der_vs_state_mlx(
  nn_name,
  min_state = NULL,
  max_state = NULL,
  inputs = NULL,
  est_parms = NULL,
  mlx_file = NULL,
  length_out = 100,
  time_nn = FALSE,
  act = "ReLU",
  beta = 20,
  transform = NULL
)

Arguments

Details

Either est_parms or mlx_file must be given. If both arguments are given, est_parms is prioritized.

Value

Dataframe with columns for the state and the corresponding derivatives

Author(s)

Dominic Bräm

Generate Derivative versus State (nlmixr2)

Description

This functions allows to generate derivative versus state data for a neural network from a NODE in nlmixr2.

Usage

der_vs_state_nlmixr(
  nn_name,
  min_state = NULL,
  max_state = NULL,
  inputs = NULL,
  est_parms = NULL,
  fit_obj = NULL,
  length_out = 100,
  time_nn = FALSE,
  act = "ReLU",
  beta = 20
)

Arguments

Details

Either est_parms or fit_obj must be given. If both arguments are given, est_parms is prioritized.

Value

Dataframe with columns for the state and the corresponding derivatives

Author(s)

Dominic Bräm

Generate Derivative versus State (NONMEM)

Description

This functions allows to generate derivative versus state data for a neural network from a NODE in NONMEM. Can also be used for nlmixr2.

Usage

der_vs_state_nm(
  nn_name,
  min_state = NULL,
  max_state = NULL,
  inputs = NULL,
  est_parms = NULL,
  nm_res_file = NULL,
  length_out = 100,
  time_nn = FALSE,
  act = "ReLU",
  beta = 20,
  transform = NULL
)

Arguments

Details

Either est_parms or nm_res_file must be given. If both arguments are given, est_parms is prioritized.

Value

Dataframe with columns for the state and the corresponding derivatives

Author(s)

Dominic Bräm

Internal: Calculate the derivatives from a NN in Monolix

Description

Calculate the derivatives from a NN for derivative versus state plots

Usage

derivative_calc_mlx(
  nn_name,
  parms,
  inputs,
  n_hidden = 5,
  time_nn = FALSE,
  act = "ReLU",
  beta = 20
)

Arguments

Value

A vector of derivatives of the NN for the state values

Author(s)

Dominic Bräm

Internal: Calculate the derivatives from a NN in NONMEM

Description

Calculate the derivatives from a NN for derivative versus state plots. Can also be used for nlmixr2.

Usage

derivative_calc_nm(
  nn_name,
  parms,
  inputs,
  n_hidden = 5,
  time_nn = FALSE,
  act = "ReLU",
  beta = 20
)

Arguments

Value

A vector of derivatives of the NN for the state values

Author(s)

Dominic Bräm

Finde path to NONMEM nmfe file

Description

To run a NONMEM model, a NONMEM nmfeXX file is required, with XX the NONMEM version. When opening the NONMEM command prompt, working directory is usually set to folder, where the nmfe file is located. When running NONMEM from R (with the run_nm function), the path and the nmfe file must be provided (as the nm_path argument). To facilitate the search for the nmfe file, this function can be used.

Usage

find_nmfe(root = "C:/")

Arguments

Details

This function assumes that the path to the nmfe file is "root/nmXXXX/run/nmfeXX, with XXXX as the NONMEM version. If any special installation settings were applied, this function might not be working.

Value

Path and name of NONMEM nmfe file, that can directly be used as nm_path argument in the run_nm function.

Author(s)

Dominic Bräm

Examples

## Not run: 
nmfe_path <- find_nmfe()
run_nm(ctl_file="./test/nm_test.ctl",nm_path=nmfe_path,create_dir=FALSE)

## End(Not run)

List of examples available

Description

Get a list of examples available in this package

Usage

get_example_list(pkg_name = "pmxNODE")

Arguments

Details

• Example 1: PK with IV drug administration

• Example 2: PK with IV drug administration

• Example 3: PK with PO drug administration

• Example 4: PK with IV drug administration and PD

Value

A list of all examples available

Author(s)

Dominic Bräm

Examples

example_list <- get_example_list()

Generate Derivative versus State Plot for individual parameter estimates (Monolix)

Description

This functions allows to generate a derivative versus state plot for a neural network from a NODE in Monolix with individual parameter estimates (EBEs).

Usage

ind_der_state_plot_mlx(
  nn_name,
  min_state = NULL,
  max_state = NULL,
  inputs = NULL,
  est_parms = NULL,
  mlx_file = NULL,
  time_nn = FALSE,
  act = "ReLU",
  ribbon = TRUE,
  length_out = 100,
  beta = 20,
  transform = NULL
)

Arguments

Details

Either est_parms or mlx_file must be given. If both arguments are given, est_parms is prioritized.

Value

Displaying derivative versus state plot

Author(s)

Dominic Bräm

Examples

mlx_path <- system.file("extdata","mlx_example1_ind.mlxtran",package="pmxNODE")
der_state_plot <- ind_der_state_plot_mlx(nn="c",
                                         min_state=0,max_state=10,
                                         mlx_file=mlx_path)

Generate Derivative versus State Plot for individual parameter estimates (nlmixr2)

Description

This functions allows to generate a derivative versus state plot for a neural network from a NODE in nlmixr2 with individual parameter estimates (EBEs).

Usage

ind_der_state_plot_nlmixr(
  nn_name,
  min_state = NULL,
  max_state = NULL,
  inputs = NULL,
  est_parms = NULL,
  fit_obj = NULL,
  length_out = 100,
  time_nn = FALSE,
  ribbon = TRUE,
  act = "ReLU",
  beta = 20
)

Arguments

Details

Either est_parms or fit_obj must be given. If both arguments are given, est_parms is prioritized.

Value

Displaying derivative versus state plot

Author(s)

Dominic Bräm

Examples

## Not run: 
ind_fit <- nlmxir2(node_model_ind,data=data,est="saem")
der_state_plot <- ind_der_state_plot_nlmixr(nn="c",min_state=0,max_state=10,
                                            fit_obj=ind_fit)

## End(Not run)

Generate Derivative versus State Plot for individual parameter estimates (NONMEM)

Description

This functions allows to generate a derivative versus state plot for a neural network from a NODE in NONMEM with individual parameter estimates (EBEs).

Usage

ind_der_state_plot_nm(
  nn_name,
  min_state = NULL,
  max_state = NULL,
  inputs = NULL,
  est_parms = NULL,
  nm_res_file = NULL,
  nm_phi_file = NULL,
  length_out = 100,
  time_nn = FALSE,
  ribbon = TRUE,
  act = "ReLU",
  beta = 20,
  transform = NULL
)

Arguments

Details

Either est_parms or nm_res_file and nm_phi_file must be given. If both arguments are given, est_parms is prioritized.

Value

Displaying derivative versus state plot

Author(s)

Dominic Bräm

Examples

res_path <- system.file("extdata","nm_example1_model_converted_ind.res",package="pmxNODE")
phi_path <- system.file("extdata","nm_example1_model_converted_ind.phi",package="pmxNODE")
der_state_plot <- ind_der_state_plot_nm(nn="c",min_state=0,max_state=10,
                                        nm_res_file=res_path,
                                        nm_phi_file=phi_path)

Generate Derivative versus State with individual parameters (Monolix)

Description

This functions allows to generate derivative versus state data for a neural network from a NODE in Monolix with individual parameters.

Usage

ind_der_vs_state_mlx(
  nn_name,
  min_state = NULL,
  max_state = NULL,
  inputs = NULL,
  est_parms = NULL,
  mlx_file = NULL,
  time_nn = FALSE,
  length_out = 100,
  act = "ReLU",
  beta = 20,
  transform = NULL
)

Arguments

Details

Either est_parms or mlx_file must be given. If both arguments are given, est_parms is prioritized.

Value

Dataframe with columns for the state and the corresponding individual derivatives

Author(s)

Dominic Bräm

Generate Derivative versus State with individual parameters (nlmixr)

Description

This functions allows to generate derivative versus state data for a neural network from a NODE in nlmixr2 with individual parameters.

Usage

ind_der_vs_state_nlmixr(
  nn_name,
  min_state = NULL,
  max_state = NULL,
  inputs = NULL,
  est_parms = NULL,
  fit_obj = NULL,
  length_out = 100,
  time_nn = FALSE,
  act = "ReLU",
  beta = 20
)

Arguments

Details

Either est_parms or nm_res_file and nm_phi_file must be given. If both arguments are given, est_parms is prioritized.

Value

Dataframe with columns for the state and the corresponding individual derivatives

Author(s)

Dominic Bräm

Generate Derivative versus State with individual parameters (NONMEM)

Description

This functions allows to generate derivative versus state data for a neural network from a NODE in NONMEM with individual parameters.

Usage

ind_der_vs_state_nm(
  nn_name,
  min_state = NULL,
  max_state = NULL,
  inputs = NULL,
  est_parms = NULL,
  nm_res_file = NULL,
  nm_phi_file = NULL,
  length_out = 100,
  time_nn = FALSE,
  act = "ReLU",
  beta = 20,
  transform = NULL
)

Arguments

Details

Either est_parms or nm_res_file and nm_phi_file must be given. If both arguments are given, est_parms is prioritized.

Value

Dataframe with columns for the state and the corresponding individual derivatives

Author(s)

Dominic Bräm

Generate individual Right-hand side data (Monolix)

Description

This functions allows to generate right-hand side data for multiple individuals with individual parameter sets, i.e., combined derivative data of multiple NNs and base-R operations.

Usage

ind_rhs_calc_mlx(
  rhs,
  inputs,
  group,
  est_parms = NULL,
  mlx_file = NULL,
  time_nn = NULL,
  act = NULL,
  beta = 20
)

Arguments

Details

Either est_parms or mlx_file must be given. If both arguments are given, est_parms is prioritized.

Value

Dataframe with columns for the inputs and the combined right-hand side data.

Author(s)

Dominic Bräm

Generate individual Right-hand side data (nlmixr2)

Description

This functions allows to generate right-hand side data for multiple individuals with individual parameter sets, i.e., combined derivative data of multiple NNs and base-R operations.

Usage

ind_rhs_calc_nlmixr(
  rhs,
  inputs,
  group,
  est_parms = NULL,
  fit_obj = NULL,
  time_nn = NULL,
  act = NULL,
  beta = 20
)

Arguments

Details

Either est_parms or fit_obj must be given. If both arguments are given, est_parms is prioritized.

Value

Dataframe with columns for the inputs and the combined right-hand side data.

Author(s)

Dominic Bräm

Generate individual Right-hand side data (NONMEM)

Description

This functions allows to generate right-hand side data for multiple individuals with individual parameter sets, i.e., combined derivative data of multiple NNs and base-R operations.

Usage

ind_rhs_calc_nm(
  rhs,
  inputs,
  group,
  est_parms = NULL,
  nm_res_file = NULL,
  nm_phi_file = NULL,
  time_nn = NULL,
  act = NULL,
  beta = 20
)

Arguments

Details

Either est_parms or nm_res_file and nm_phi_file must be given. If both arguments are given, est_parms is prioritized.

Value

Dataframe with columns for the inputs and the combined right-hand side data.

Author(s)

Dominic Bräm

Generate individual Right-hand side data plot (Monolix)

Description

This functions allows to generate a right-hand side plot with multiple subjects, i.e., combined derivative data of multiple NNs and base-R operations.

Usage

ind_rhs_plot_mlx(
  rhs,
  x_var,
  inputs,
  group,
  est_parms = NULL,
  mlx_file = NULL,
  time_nn = NULL,
  act = NULL,
  beta = 20
)

Arguments

Details

Either est_parms or mlx_file must be given. If both arguments are given, est_parms is prioritized.

Value

ggplot of right-hand side plot for all individuals

Author(s)

Dominic Bräm

Examples

# Generate individual rhs-plot for predicted observations

mlx_path <- system.file("extdata","mlx_example1_ind.mlxtran",package="pmxNODE")
data_path <- system.file("extdata","mlx_example1_ind","predictions.txt",package="pmxNODE")

est_parms <- indparm_extractor_mlx(mlx_path)

input_data <- read.table(data_path,sep=",",header=TRUE)[,c("id","indivPred_mode","time")]
colnames(input_data) <- c("id","NNc","NNct")
                  
rhs_plot <- ind_rhs_plot_mlx(rhs="NNc + NNct",
                             x_var = "NNc",
                             group = "id",
                             inputs = input_data,
                             est_parms = est_parms,
                             time_nn = c(FALSE, TRUE))

Generate individual Right-hand side data plot (nlmixr2)

Description

This functions allows to generate a right-hand side plot with multiple subjects, i.e., combined derivative data of multiple NNs and base-R operations.

Usage

ind_rhs_plot_nlmixr(
  rhs,
  x_var,
  inputs,
  group,
  est_parms = NULL,
  fit_obj = NULL,
  time_nn = NULL,
  act = NULL,
  beta = 20
)

Arguments

Details

Either est_parms or fit_obj must be given. If both arguments are given, est_parms is prioritized.

Value

ggplot of right-hand side for all individuals.

Author(s)

Dominic Bräm

Examples

## Not run: 
ind_fit <- nlmxir2(node_model_ind,data=data,est="saem")
rhs_plot <- ind_rhs_plot_mlx(rhs="NNc + NNct",
                             x_var = "NNc",
                             group = "id",
                             inputs = input_data,
                             fit_obj = ind_fit)

## End(Not run)

Generate individual Right-hand side data plot (NONMEM)

Description

This functions allows to generate a right-hand side plot with multiple subjects, i.e., combined derivative data of multiple NNs and base-R operations.

Usage

ind_rhs_plot_nm(
  rhs,
  x_var,
  inputs,
  group,
  est_parms = NULL,
  nm_res_file = NULL,
  nm_phi_file = NULL,
  time_nn = NULL,
  act = NULL,
  beta = 20
)

Arguments

Details

Either est_parms or mlx_file must be given. If both arguments are given, est_parms is prioritized.

Value

ggplot of right-hand side plot for all individuals

Author(s)

Dominic Bräm

Examples

# Generate individual rhs-plot for predicted observations

res_path <- system.file("extdata","nm_example1_model_converted_ind.res",package="pmxNODE")
phi_path <- system.file("extdata","nm_example1_model_converted_ind.phi",package="pmxNODE")
data_path <- system.file("extdata","nm_example1.tab",package="pmxNODE")

est_parms <- indparm_extractor_nm(res_path,phi_path)

input_data <- read.table(data_path,skip=1,header=TRUE)[,c("ID","IPRED","TIME")]
colnames(input_data) <- c("ID","NNc","NNt")
                  
rhs_plot <- ind_rhs_plot_nm(rhs="NNc + NNt",
                             x_var = "NNc",
                             inputs = input_data,
                             group = "ID",
                             est_parms=est_parms,
                             time_nn = c(FALSE, TRUE))

Monolix individual estimations extractor

Description

When the Monolix model has been run, this function allows to extract the estimated individual parameters (EBEs) from the Monolix run folder.

Usage

indparm_extractor_mlx(model_name)

Arguments

Value

Data frame with individual parameter estimates (EBEs)

Author(s)

Dominic Bräm

Examples

mlx_path <- system.file("extdata","mlx_example1_ind.mlxtran",package="pmxNODE")
est_parms <- indparm_extractor_mlx(mlx_path)

nlmixr individual estimations extractor

Description

When the nlmixr model has been run, this function allows to extract the estimated individual parameters for NN parameters by combining fixed effects and random effects

Usage

indparm_extractor_nlmixr(fit_obj)

Arguments

Value

Data frame with individual parameter estimates for NN parameters

Author(s)

Dominic Bräm

Examples

## Not run: 
fit_ind <- nlmixr(model_with_iiv,data=data,est="saem")
est_parms <- indparm_extractor_nlmixr(fit_ind)

## End(Not run)

NONMEM individual estimations extractor

Description

When the NOMEM model has been run, this function allows to extract the estimated individual parameters for NN parameters by combining information from the .res and the .phi file

Usage

indparm_extractor_nm(res_file, phi_file)

Arguments

Value

Data frame with individual parameter estimates for NN parameters

Author(s)

Dominic Bräm

Examples

res_path <- system.file("extdata","nm_example1_model_converted_ind.res",package="pmxNODE")
phi_path <- system.file("extdata","nm_example1_model_converted_ind.phi",package="pmxNODE")
est_parms <- indparm_extractor_nm(res_path,phi_path)

Internal: Monolix file initializer

Description

Initializes a .mlxtran file based on a converted model file, data, and initial values

Usage

mlx_model_initializer(
  model_name,
  model_file,
  data_file,
  header_types,
  parm_names,
  parm_inis,
  theta_names,
  theta_inis,
  pop = FALSE,
  omega_inis = NULL,
  pre_fixef = NULL,
  obs_types = NULL,
  mapping = NULL
)

Arguments

Details

If no specific file name was given to nn_converter_mlx through the mlx_name argumen, the file is standardized to name of model file_mlx_file_pop/ind where pop if pop=TRUE or ind if pop=FALSE in nn_converter_mlx.

Value

No return value, saving a Monolix .mlxtran file.

Author(s)

Dominic Bräm

Internal: Extract model parameters

Description

Extract model parameters and corresponding initial values from the model input line

Usage

model_parm_extractor_mlx(text)

Arguments

Value

• [1] - list of parameter names

• [2] - list of initial values

Author(s)

Dominic Bräm

Internal: Update input line

Description

Update a Monolix model input to define non-NN and NN parameters as model inputs

Usage

model_parm_updater_mlx(text, model_parm_names, nn_thetas)

Arguments

Value

The Monolix model including all non-NN and NN parameters in input = ...

Author(s)

Dominic Bräm

Internal: Extract activation function of a NN

Description

Extract the activation function to be used for a NN given as the act argument in the NN function

Usage

nn_act_extractor(text)

Arguments

Value

Activation function to be used for a NN

Author(s)

Dominic Bräm

NN converter for Monolix

Description

This function converts a Monolix model file that includes pseudo-functions for NNs as described in Details into a model that can be used in Monolix. An example Monolix model can be opened with the function open_mlx_example(). In addition, it allows to generate a Monolix .mlxtran file with automatically initialized parameters and estimation settings.

Usage

nn_converter_mlx(
  mlx_path,
  pop_only = FALSE,
  theta_scale = 0.1,
  eta_scale = 0.1,
  pre_fixef = NULL,
  gen_mlx_file = FALSE,
  mlx_name = NULL,
  data_file = NULL,
  header_types = NULL,
  obs_types = NULL,
  mapping = NULL,
  seed = 1908
)

Arguments

Details

An example of model file could look like following

DESCRIPTION:

A Monolix model for conversion

[LONGITUDINAL]

input = {V=2}

PK:

depot(target=C)

EQUATION:

ddt_C = NN1(state=C,min_init=1,max_init=300) +

                amtDose * NN2(state=t,min_init=0.5,max_init=50,time_nn=TRUE)

Cc = C/V

OUTPUT:

output = Cc

• Note that the parameters in the input need to have an initial value

• NN functions need to be of form NNX(...) where X is the name of the NN so references between the same NN, e.g., as output of absorption compartment and input to central compartment, can be made. Arguments to NNX are

  • state= defines the state to be used in the NN. For time, use t.

  • min_init= defines the minimal activation point for the NN, i.e., minimal expected state

  • max_init= defines the maximal activation point for the NN, i.e., maximal expected state

  • n_hidden= (optional) defines the number of neurons in the hidden layer, default is 5

  • act= (optional) defines activation function in the hidden layer, ReLU and Softplus implemented, default is ReLU

  • time_nn= (optional) defines whether the NN should be assumed to be a time-dependent NN and consequently all weights from input to hidden layer should be strictly negative.

Note: Converted Monolix model file will be saved under unconverted_file_converted.txt

Value

Saving a converted Monolix model file under mlx_path_converted.txt and optionally a Monolix file (mlx_name.mlxtran) if gen_mlx_file=TRUE

Author(s)

Dominic Bräm

Examples

## Not run: 
nn_converter_mlx("mlx_model2.txt",
                 pop_only=TRUE,gen_mlx_file=TRUE,
                 data_file="TMDD_dataset.csv",
                 header_types=c("id","time","amount","observation"))

est_parms <- pre_fixef_extractor_mlx("mlx_model2_time_nn_mlx_file_pop.mlxtran")

nn_converter_mlx("mlx_model2.txt",
                 pop_only=FALSE,gen_mlx_file=TRUE,
                 data_file="TMDD_dataset.csv",
                 header_types=c("id","time","amount","observation"),
                 pre_fixef=est_parms)

## End(Not run)

NN converter for nlmixr

Description

This function converts a nlmixr model function that includes pseudo-functions for NNs as described in Details into a model function that can be used in the nlmixr function.

Usage

nn_converter_nlmixr(
  f_ode,
  pop_only = FALSE,
  theta_scale = 0.1,
  eta_scale = 0.1,
  pre_fixef = NULL,
  seed = 1908
)

Arguments

Details

• state= defines the state to be used in the NN. For time, use t.

• min_init= defines the minimal activation point for the NN, i.e., minimal expected state

• max_init= defines the maximal activation point for the NN, i.e., maximal expected state

• n_hidden= (optional) defines the number of neurons in the hidden layer, default is 5

• act= (optional) defines activation function in the hidden layer, ReLU and Softplus implemented, default is ReLU

• time_nn= (optional) defines whether the NN should be assumed to be a time-dependent NN and consequently all weights from input to hidden layer should be strictly negative.

Value

A converted nlmixr model function

Author(s)

Dominic Bräm

Examples

## Not run: 
f_ode_pop <- function(){
ini({
  lV <- 1
  prop.err <- 0.1
})
model({
  V <- lV
  d/dt(centr)  =  NN1(state=centr,min_init=0,max_init=300)
  cp = centr / V
  cp ~ prop(prop.err)
  })
}

f_new_pop <- nn_converter_nlmixr(f_ode_pop,pop_only = TRUE)

fit_pop <- nlmixr2(f_new_pop,data,est="bobyqa")

f_ode <- function(){
  ini({
    lV <- 1
    eta.V ~ 0.1
    prop.err <- 0.1
  })
  model({
    V <- lV * exp(eta.V)
    d/dt(centr)  =  NN1(state=centr,min_init=0,max_init=300)
    cp = centr / V
    cp ~ prop(prop.err)
  })
}

f_new <- nn_converter_nlmixr(f_ode,pop_only = FALSE, pre_fixef = fit_pop$fixef)

## End(Not run)

NN converter for NONMEM

Description

This function converts a NONMEM model file that includes pseudo-functions for NNs as described in Details into a model that can be used in NONMEM. An example NONMEM model can be opened with the function open_nm_example().

Usage

nn_converter_nm(
  ctl_path,
  pop_only = FALSE,
  theta_scale = 0.1,
  eta_scale = 0.001,
  pre_fixef = NULL,
  seed = 1908
)

Arguments

Details

An example of model file could look like following

$SIZES LVR=80 LNP4=40000

$PROB RUN

$INPUT C ID TIME AMT DV DOSE EVID

$DATA data_example1_nm.csv IGNORE=C

$SUBROUTINES ADVAN13

$MODEL

COMP(Centr)

$PK

lV = THETA(1)

V = lV * EXP(ETA(1))

$DES

DADT(1) = NNc(state=A(1),min_init=0.5,max_init=5) +

               DOSE * NNt(state=T,min_init=1,max_init=5,time_nn=TRUE)

$ERROR

Cc = A(1)/V

Y=Cc*(1+EPS(1)) + EPS(2)

$THETA

2 ; [V]

$OMEGA

0.1 ; [V]

$SIGMA

0.1

0.1

$ESTIMATION METHOD=1 MAXEVAL=9999 INTER PRINT=5

$TABLE ID TIME DV IPRED=CIPRED AMT NOPRINT FILE=nm_example1.tab

• Note that size of problem should be increased, as in the model above with $SIZES LVR=80 LNP4=40000

• NN functions need to be of form NNX(...) where X is the name of the NN so references between the same NN, e.g., as output of absorption compartment and input to central compartment, can be made. Arguments to NNX are

  • state= defines the state to be used in the NN. For time, use t.

  • min_init= defines the minimal activation point for the NN, i.e., minimal expected state

  • max_init= defines the maximal activation point for the NN, i.e., maximal expected state

  • n_hidden= (optional) defines the number of neurons in the hidden layer, default is 5

  • act= (optional) defines activation function in the hidden layer, ReLU and Softplus implemented, default is ReLU

  • time_nn= (optional) defines whether the NN should be assumed to be a time-dependent NN and consequently all weights from input to hidden layer should be strictly negative.

Note: Converted NONMEM model file will be saved under unconverted_file_converted.ctl

Value

Saving a converted NONMEM model file under ctl_path_converted.ctl

Author(s)

Dominic Bräm

Examples

## Not run: 
nn_converter_nm("nm_example_model.ctl",pop_only = TRUE)

est_parms <- pre_fixef_extractor_nm("nm_example_model_converted.res")

nn_converter_nm("nm_example_model.ctl",pop_only = FALSE,pre_fixef=est_parms)

## End(Not run)

Internal: Error for incomplete NNs

Description

Raises error if one or more NNs are not complete, i.e., not all required arguments are given

Usage

nn_errors(nn_tests)

Arguments

Details

Currently required arguments are: NN-name NNX(), state (state=), minimal (min_init=) and maximal (max_init=) activation points

Value

Error: Missing argument; NNX: missing arguments

Author(s)

Dominic Bräm

Internal: Define NN ETAs in NONMEM

Description

Define random effects of NN parameters as ETAs used in the $PK section in NONMEM

Usage

nn_eta_def_nm(number, eta_start, n_hidden = 5, time_nn = FALSE)

Arguments

Details

Parameter definition in form of etaW = ETA(X) with X the next number of ETA
e.g., if
etaV = ETA(1)
etakel = ETA(2)
X is equal to 3

Value

Vector with all NN parameter ETA definitions for a NN

Author(s)

Dominic Bräm

Internal: Initialize random effects on NN parameters in nlmixr

Description

Define the standard deviation of random effects on NN parameters for the ini section of the nlmixr model

Usage

nn_eta_initializer_nlmixr(
  number,
  n_hidden = 5,
  eta_scale = 0.1,
  time_nn = FALSE
)

Arguments

Value

Vector of initial random effects on NN parameters for one specific NN

Author(s)

Dominic Bräm

Internal: Set initial ETA estimates in NONMEM

Description

Set the initial ETA estimates for the $OMEGA block in the NONMEM model file

Usage

nn_eta_initializer_nm(
  number,
  theta_inis,
  pop = FALSE,
  n_hidden = 5,
  eta_scale = 0.1,
  time_nn = FALSE
)

Arguments

Details

ETAs are fixed to 0 if

• population fit is enabled through pop=TRUE, all ETAs are fixed to 0

• the corresponding THETA is fixed to 0 due to non-activity of the neuron

Value

Vector of initial NN parameter ETA values for one specific NN

Author(s)

Dominic Bräm

Internal: Extract NN function from model

Description

Extract the entire NN functions from a model file

Usage

nn_extractor(text)

Arguments

Value

NN function as string in form of "NN1(state=t,min_init=1,max_init=5)"

Author(s)

Dominic Bräm

Internal: Generate NN code for Monolix

Description

Generate the explicit code for a NN in Monolix

Usage

nn_generator_mlx(
  number,
  state,
  n_hidden = 5,
  act = "ReLU",
  time_nn = FALSE,
  beta = 20
)

Arguments

Details

Structure of one unit in the hidden layer:
h1 = max(0,W1 * state + b1)
h2 = max(0,W2 * state + b2)
...
NN = h1 + h2 + ...

Value

Explicit NN code in Monolix as list of lines in Monolix model file

Author(s)

Dominic Bräm

Internal: Generate NN code for nlmixr

Description

Generate the explicit code for a NN in nlmixr

Usage

nn_generator_nlmixr(
  number,
  state,
  n_hidden = 5,
  act = "ReLU",
  time_nn = FALSE,
  beta = 20
)

Arguments

Details

Structure of one unit in the hidden layer:
h1 = W1 * state + b1
if(h1 < 0) {h1 <- 0}
h2 = W2 * state + b2
...
NN = h1 + h2 + ...

Value

Explicit NN code in nlmixr as list of lines in nlmixr model file

Author(s)

Dominic Bräm

Internal: Generate NN code for NONMEM

Description

Generate the explicit code for a NN in NONMEM

Usage

nn_generator_nm(
  number,
  state,
  n_hidden = 5,
  act = "ReLU",
  time_nn = FALSE,
  beta = 20
)

Arguments

Details

Structure of one unit in the hidden layer:
h1 = 0
h1_thres = W1 * state + b1
IF (h1_thres.GT.h1) h1 = h1_thres
h2 = 0
...
NN = h1 + h2 + ...

Value

Explicit NN code in NONMEM as list of lines in NONMEM model file

Author(s)

Dominic Bräm

Internal: Extract maximal activation point of NN

Description

Extract the maximal activation point of a NN given as the max_init argument in the NN function

Usage

nn_maxini_extractor(text)

Arguments

Value

Maximal activation point for a NN

Author(s)

Dominic Bräm

Internal: Extract minimal activation point of NN

Description

Extract the minimal activation point of a NN given as the min_init argument in the NN function

Usage

nn_minini_extractor(text)

Arguments

Value

Maximal activation point for a NN

Author(s)

Dominic Bräm

Internal: Extracting number of units in hidden layer

Description

Extracts the number of units in the hidden layer of a NN given as the n_hidden argument in the NN function

Usage

nn_nhidden_extractor(text)

Arguments

Details

Not yet in use

Value

Number of units in the hidden layer for a specific NN

Author(s)

Dominic Bräm

Internal: Extract name of a NN

Description

Extract the name of a NN given as the NNX argument

Usage

nn_number_extractor(text)

Arguments

Details

Potentially be moved to argument in NN function in form of NN(name=1,state=C,min_init=1,max_init=5) at a later stage

Value

Name of the NN

Author(s)

Dominic Bräm

Internal: Definition of NN parameters in nlmixr

Description

Define NN parameters consisting of typical parameter and potentially random effects in the model section of a nlmixr model

Usage

nn_parm_setter_nlmixr(
  number,
  pop = FALSE,
  n_hidden = 5,
  eta_model = c("prop", "add"),
  time_nn = FALSE
)

Arguments

Details

eta_model is currently set to proportional as previous investigations showed better stability of fit with this setting

Value

List of parameter definition to be used in the model section of the nlmixr model

Author(s)

Dominic Bräm

Internal: Definition of NN parameters in NONMEM

Description

Define NN parameters consisting of typical parameter and potentially random effects in the $PK section

Usage

nn_parm_setter_nm(
  number,
  pop = FALSE,
  n_hidden = 5,
  eta_model = c("prop", "add"),
  time_nn = FALSE
)

Arguments

Details

eta_model is currently set to proportional as previous investigations showed better stability of fit with this setting in NONMEM

Value

List of parameter definition to be used in the $PK section of the NONMEM model

Author(s)

Dominic Bräm

Internal: Reduces NN function in model file

Description

Converts NN functions of form NN1(state=C,min_init=1,max_init=5) to model usable form of NN1.

Usage

nn_reducer(text)

Arguments

Value

Converted model with NN1(state=C,min_init=1,max_init=5) expressed as NN1

Author(s)

Dominic Bräm

Internal: Extract state of a NN

Description

Extract the state to be used for a NN given as the state argument in the NN function

Usage

nn_state_extractor(text)

Arguments

Value

State to be used for a NN

Author(s)

Dominic Bräm

Internal: Test NNs for completeness

Description

Tests whether all required arguments are given to each NN

Usage

nn_tester(text)

Arguments

Details

Currently required arguments are: NN-name NNX(), state (state=), minimal (min_init=) and maximal (max_init=) activation points

Value

Missing arguments of a NN, “None” if all arguments are given

Author(s)

Dominic Bräm

Internal: Define NN parameters names

Description

Get parameter names for one specific NN in the model, i.e., weight and bias names

Usage

nn_theta_def_mlx(number, n_hidden = 5, time_nn = FALSE)

Arguments

Value

Vector with all NN parameter names

Author(s)

Dominic Bräm

Internal: Define NN THETAs in NONMEM

Description

Define NN THETAS used in the $PK section in NONMEM

Usage

nn_theta_def_nm(number, theta_start, n_hidden = 5, time_nn = FALSE)

Arguments

Details

Parameter definition in form of lW = THETA(X) with X the next number of THETA
e.g., if
lV = THETA(1)
lkel = THETA(2)
X is equal to 3

Value

Vector with all NN parameter THETA definitions for a NN

Author(s)

Dominic Bräm

Internal: Calculate initial NN parameter values

Description

Calculate the initial NN parameter values, such that activation points are within the range between min_init and max_init defined in the un-converter Monolix model file

Usage

nn_theta_initializer_mlx(
  number,
  xmini,
  xmaxi,
  n_hidden = 5,
  theta_scale = 0.1,
  pre_fixef = NULL,
  time_nn = FALSE,
  act = "ReLU",
  beta = 20
)

Arguments

Details

• theta_scale is the scale in which the weights from input to hidden layer are initialized, i.e., 0.1 initializes weights between -0.3 and 0.3; 0.01 initializes weights between -0.03 and 0.03

• time_nn defines whether the NN is a time-dependent NN with the restriction that all weights from input to hidden layer are negative

Value

Vector of initial NN parameter values for one specific NN

Author(s)

Dominic Bräm

Internal: Initialize typical NN parameter in nlmixr

Description

Calculate the initial typical NN parameter values, such that activation points are within the range between min_init and max_init defined in the un-converter NONMEM model file, and define the the typical NN parameters for the ini section of the nlmixr model

Usage

nn_theta_initializer_nlmixr(
  number,
  xmini,
  xmaxi,
  n_hidden = 5,
  theta_scale = 0.1,
  pre_fixef = NULL,
  time_nn = FALSE,
  act = "ReLU",
  beta = 20
)

Arguments

Details

• theta_scale is the scale in which the weights from input to hidden layer are initialized, i.e., 0.1 initializes weights between -0.3 and 0.3; 0.01 initializes weights between -0.03 and 0.03

• time_nn defines whether the NN is a time-dependent NN with the restriction that all weights from input to hidden layer are negative

Value

Vector of initial typical NN parameters for one specific NN

Author(s)

Dominic Bräm

Internal: Calculate initial NN parameter values in NONMEM

Description

Calculate the initial NN parameter values, such that activation points are within the range between min_init and max_init defined in the un-converter NONMEM model file

Usage

nn_theta_initializer_nm(
  number,
  xmini,
  xmaxi,
  n_hidden = 5,
  theta_scale = 0.1,
  pre_fixef = NULL,
  time_nn = FALSE,
  act = "ReLU",
  beta = 20
)

Arguments

Details

• theta_scale is the scale in which the weights from input to hidden layer are initialized, i.e., 0.1 initializes weights between -0.3 and 0.3; 0.01 initializes weights between -0.03 and 0.03

• time_nn defines whether the NN is a time-dependent NN with the restriction that all weights from input to hidden layer are negative

Value

Vector of initial NN parameter THETA values for one specific NN

Author(s)

Dominic Bräm

Internal: Extracting time-NN argument

Description

Extracts, whether a NN should be treated as time-dependent NN

Usage

nn_time_nn_extractor(text)

Arguments

Details

Time-dependent NNs have different model structures, i.e., weights from input to hidden layer are set to negative through w'=-w^2

Value

List of boolean expression whether NN should be treated as time-NN (TRUE) or not (FALSE)

Author(s)

Dominic Bräm

Open Monolix example model

Description

Opens unconverted Monolix example model with NN functions

Usage

open_mlx_example(model_nr = 1)

Arguments

Details

Example numbers 1-4 currently available.

Value

Returns example model text.

Author(s)

Dominic Bräm

Open NONMEM example model

Description

Opens unconverted NONMEM example model with NN functions

Usage

open_nm_example(model_nr = 1)

Arguments

Details

Example numbers 1-4 currently available.

Value

Returns example model text.

Author(s)

Dominic Bräm

Monolix estimations extractor

Description

When the Monolix model has been run, e.g., with only population estimation, this function allows to extract the estimated parameters from the Monolix run folder. This function is meant, e.g., to get initial values for a Monolix run with inter-individual variability and to be then used as pre_fixef argument in the nn_converter_mlx function

Usage

pre_fixef_extractor_mlx(model_name)

Arguments

Value

Named vector of Monolix parameter estimations

Author(s)

Dominic Bräm

Examples

mlx_path <- system.file("extdata","mlx_example1_ind.mlxtran",package="pmxNODE")
est_parms <- pre_fixef_extractor_mlx(mlx_path)

THETA extraction from results file

Description

Function to extract THETA estimates from a results file of an already run NONMEM file.

Usage

pre_fixef_extractor_nm(res_path)

Arguments

Details

Can be used, e.g., to initialize THETAs of a run with inter-individual variability with estimated THETAs of a previous population run without inter-individual variability. Parameters, for which final gradient is equal to 0 are fixed to 0, because a gradient of 0 indicates that corresponding neuron was inactivated during parameter estimation.

Value

Named vector with parameter estimates from the previous run

Author(s)

Dominic Bräm

Examples

res_path <- system.file("extdata","nm_example1_model_converted_ind.res",package="pmxNODE")
pre_fixef <- pre_fixef_extractor_nm(res_path)

Generate Right-hand side data (Monolix)

Description

This functions allows to generate right-hand side data, i.e., combined derivative data of multiple NNs and base-R operations.

Usage

rhs_calc_mlx(
  rhs,
  inputs,
  est_parms = NULL,
  mlx_file = NULL,
  time_nn = NULL,
  act = NULL,
  beta = 20
)

Arguments

Details

Either est_parms or mlx_file must be given. If both arguments are given, est_parms is prioritized.

Value

Dataframe with columns for the inputs and the combined right-hand side data.

Author(s)

Dominic Bräm

Generate Right-hand side data (nlmixr2)

Description

This functions allows to generate right-hand side data, i.e., combined derivative data of multiple NNs and base-R operations.

Usage

rhs_calc_nlmixr(
  rhs,
  inputs,
  est_parms = NULL,
  fit_obj = NULL,
  time_nn = NULL,
  act = NULL,
  beta = 20
)

Arguments

Details

Either est_parms or fit_obj must be given. If both arguments are given, est_parms is prioritized.

Value

Dataframe with columns for the inputs and the combined right-hand side data.

Author(s)

Dominic Bräm

Generate Right-hand side data (NONMEM)

Description

This functions allows to generate right-hand side data, i.e., combined derivative data of multiple NNs and base-R operations.

Usage

rhs_calc_nm(
  rhs,
  inputs,
  est_parms = NULL,
  nm_res_file = NULL,
  time_nn = NULL,
  act = NULL,
  beta = 20
)

Arguments

Details

Either est_parms or nm_res_file must be given. If both arguments are given, est_parms is prioritized.

Value

Dataframe with columns for the inputs and the combined right-hand side data.

Author(s)

Dominic Bräm

Generate Right-hand side data plot (Monolix)

Description

This functions allows to generate right-hand side plot, i.e., combined derivative data of multiple NNs and base-R operations.

Usage

rhs_plot_mlx(
  rhs,
  x_var,
  inputs,
  est_parms = NULL,
  mlx_file = NULL,
  time_nn = NULL,
  act = NULL,
  beta = 20
)

Arguments

Details

Either est_parms or mlx_file must be given. If both arguments are given, est_parms is prioritized.

Value

ggplot of right-hand side data.

Author(s)

Dominic Bräm

Examples

mlx_path <- system.file("extdata","mlx_example1_ind.mlxtran",package="pmxNODE")
est_parms <- pre_fixef_extractor_mlx(mlx_path)
rhs_plot <- rhs_plot_mlx(rhs="NNc + WT * NNct",
                         x_var = "NNc",
                         inputs = data.frame(NNc = 1:100,
                                             NNct = seq(0,10,length.out=100),
                                             WT = rep(50,100)),
                         est_parms=est_parms,
                         time_nn = c(FALSE, TRUE))

Generate Right-hand side data plot (nlmixr2)

Description

This functions allows to generate right-hand side plot, i.e., combined derivative data of multiple NNs and base-R operations.

Usage

rhs_plot_nlmixr(
  rhs,
  x_var,
  inputs,
  est_parms = NULL,
  fit_obj = NULL,
  time_nn = NULL,
  act = NULL,
  beta = 20
)

Arguments

Details

Either est_parms or mlx_file must be given. If both arguments are given, est_parms is prioritized.

Value

Dataframe with columns for the inputs and the combined right-hand side data.

Author(s)

Dominic Bräm

Examples

## Not run: 
pop_fit <- nlmixr2(node_model,data=data,est="bobyqa")
rhs_plot <- rhs_plot_nlmixr(rhs="NNc + WT * NNct",
                         x_var = "NNc",
                         inputs = data.frame(NNc = 1:100,
                                             NNct = seq(0,10,length.out=100),
                                             WT = rep(50,100)),
                         est_parms=pop_fit$fixef)

## End(Not run)

Generate Right-hand side data plot (NONMEM)

Description

This functions allows to generate right-hand side plot, i.e., combined derivative data of multiple NNs and base-R operations.

Usage

rhs_plot_nm(
  rhs,
  x_var,
  inputs,
  est_parms = NULL,
  nm_res_file = NULL,
  time_nn = NULL,
  act = NULL,
  beta = 20
)

Arguments

Details

Either est_parms or nm_res_file must be given. If both arguments are given, est_parms is prioritized.

Value

ggplot of right-hand side data.

Author(s)

Dominic Bräm

Examples

res_path <- system.file("extdata","nm_example1_model_converted_ind.res",package="pmxNODE")
est_parms <- pre_fixef_extractor_nm(res_path)
rhs_plot <- rhs_plot_nm(rhs="NNc + DOSE * NNt",
                         x_var = "NNc",
                         inputs = data.frame(NNc = 1:100,
                                             NNt = seq(0,10,length.out=100),
                                             DOSE = rep(50,100)),
                         est_parms=est_parms,
                         time_nn = c(FALSE, TRUE))

Run Monolix from R

Description

Runs Monolix from R

Usage

run_mlx(mlx_file)

Arguments

Details

All paths must be given in R-style, i.e., slashes instead of backslashes. Paths can be absolute or relative.

Value

No return value, running the specified model in Monolix via lixoftConnectors.

Author(s)

Dominic Bräm

Examples

## Not run: 
run_mlx("mlx_file.mlxtran")

## End(Not run)

Run NONMEM from R

Description

Runs NONMEM from R

Usage

run_nm(
  ctl_file,
  nm_path,
  parralel_command = NULL,
  create_dir = TRUE,
  data_file = NULL
)

Arguments

Details

All paths must be given in R-style, i.e., slashes instead of backslashes. Paths can be absolute or relative.

Value

No return value, running the specified model in NONMEM via command line.

Author(s)

Dominic Bräm

Examples

## Not run: 
run_nm("./test/nm_test.ctl","c:/nm75g64/run/nmfe75",
       parralel_command = "-parafile=C:/nm75g64/run/mpiwini8.pnm [nodes]=30",
       data_file="~/Test/test/test_data.csv")

## End(Not run)

Initialize software (Suspended)

Description

Initialize the pharmacometric software you want to use (Monolix, nlmixr or NONMEM). Must be used before nn_converter functions can be used for Monolix and nlmixr.

Usage

software_initializer(
  software = c("Monolix", "nlmixr", "NONMEM"),
  mlx_path = NULL
)

Arguments

Details

For Monolix, the lixoftConnectors package is loaded. For loading, the path to the Monolix location (under Windows usually C:/ProgramData/Lixoft/MonolixSuiteXXXX with XXXX as the version) is required. Note: nlmixr2 and lixoftConnectors share function getData. If both, nlmixr and Monolix, get initialized, getData will be used from the package initialized second

Value

Initialization of software

Author(s)

Dominic Bräm

Examples

## Not run: 
software_initializer(software="NONMEM")
software_initializer(software="nlmixr")
software_initializer(software="Monolix",mlx_path="C:/ProgramData/Lixoft/MonolixSuite2021R2")

## End(Not run)

Internal: Correct NONMEM states

Description

For NNs to be handled correctly, the state argument may not include brackets. state_correcter_nm translates argument state=A(1) to state=A1 and A1=A(1)

Usage

state_correcter_nm(text)

Arguments

Value

• [1] NN with corrected state argument

• [2] State translation, e.g., “A1=A(1)”

Author(s)

Dominic Bräm