Version:	4.0.1
Title:	Facilities for Simulating from ODE-Based Models
Maintainer:	Matthew L. Fidler <matthew.fidler@gmail.com>
Depends:	R (≥ 4.0.0)
Suggests:	Matrix, DT, covr, crayon, curl, digest, dplyr (≥ 0.8.0), ggrepel, gridExtra, htmltools, knitr, learnr, microbenchmark, nlme, remotes, rlang, rmarkdown, scales, shiny, stringi, symengine, testthat, tidyr, usethis, vdiffr (≥ 1.0), withr, xgxr, pillar, tibble, units (≥ 0.6-0), rsconnect, devtools, patchwork, nlmixr2data, lifecycle, kableExtra, pmxTools
Imports:	PreciseSums (≥ 0.7), Rcpp (≥ 0.12.3), backports, cli (≥ 2.0.0), checkmate, ggplot2 (≥ 3.4.0), inline, lotri (≥ 1.0.0), magrittr, memoise, methods, rex, sys, tools, utils, dparser (≥ 1.3.1-12), rxode2ll(≥ 2.0.9), data.table (≥ 1.12.4), qs (≥ 0.26.3)
Description:	Facilities for running simulations from ordinary differential equation ('ODE') models, such as pharmacometrics and other compartmental models. A compilation manager translates the ODE model into C, compiles it, and dynamically loads the object code into R for improved computational efficiency. An event table object facilitates the specification of complex dosing regimens (optional) and sampling schedules. NB: The use of this package requires both C and Fortran compilers, for details on their use with R please see Section 6.3, Appendix A, and Appendix D in the "R Administration and Installation" manual. Also the code is mostly released under GPL. The 'VODE' and 'LSODA' are in the public domain. The information is available in the inst/COPYRIGHTS.
BugReports:	https://github.com/nlmixr2/rxode2/issues/
NeedsCompilation:	yes
VignetteBuilder:	knitr
License:	GPL (≥ 3)
URL:	https://nlmixr2.github.io/rxode2/, https://github.com/nlmixr2/rxode2/
RoxygenNote:	7.3.2
Biarch:	true
LinkingTo:	sitmo, lotri (≥ 1.0.0), PreciseSums (≥ 0.7), Rcpp, RcppArmadillo (≥ 0.9.300.2.0), BH, RcppParallel, RcppEigen (≥ 0.3.3.9.2), StanHeaders (≥ 2.21.0.7), dparser (≥ 1.3.1-12)
Encoding:	UTF-8
LazyData:	true
Language:	en-US
Config/testthat/edition:	3
Packaged:	2025-07-17 18:04:26 UTC; matt
Author:	Matthew L. Fidler [aut, cre], Wenping Wang [aut], Alan Hindmarsh [ctb], Arun Srinivasan [ctb], Awad H. Al-Mohy [ctb], Bill Denney [ctb], Cleve Moler [ctb], David Cooley [ctb], Drew Schmidt [ctb], Ernst Hairer [ctb], Gabriel Staples [ctb], Gerhard Wanner [ctb], Gilbert Stewart [ctb], Goro Fuji [ctb], Hadley Wickham [ctb], Igor Kushnir [ctb], Jack Dongarra [ctb], Jim Bunch [ctb], Kevin Ushey [ctb], Linda Petzold [ctb], Martin Maechler [ctb], Matt Dowle [ctb], Matteo Fasiolo [ctb], Melissa Hallow [aut], Michel Lang [ctb], Morwenn [ctb], Nicholas J. Higham [ctb], Omar Elashkar [ctb], Richard Upton [ctb], Roger B. Sidje [ctb], Simon Frost [ctb], Yu Feng [ctb], Zufar Mulyukov [ctb]
Repository:	CRAN
Date/Publication:	2025-07-18 00:10:02 UTC

Update Solved object with '+'

Description

Update Solved object with '+'

Usage

## S3 method for class 'rxSolve'
solved + new

Arguments

Value

new solved object

Author(s)

Matthew L. Fidler

This function asserts the requested rename makes sense

Description

It returns the new expression, old expression, new variable and old variable

Usage

.assertRenameErrorModelLine(line, vars)

Arguments

Value

list(new, old, newChar, oldChar)

Author(s)

Matthew L. Fidler

cbind Ome

Description

cbind Ome

Usage

.cbindOme(et, mat, n)

Arguments

Value

data frame with et combined with simulated omega matrix values

Author(s)

Matthew Fidler

Clear/Set pipeline

Description

Clear/Set pipeline

Usage

.clearPipe(
  rx = NULL,
  inits = NULL,
  events = NULL,
  params = NULL,
  iCov = NULL,
  keep = NULL,
  thetaMat = NULL,
  omega = NULL,
  sigma = NULL,
  dfObs = NULL,
  dfSub = NULL,
  nSub = NULL,
  nStud = NULL
)

Arguments

Value

None, clears rxode2 pipeline

Collect warnings and just warn once.

Description

Collect warnings and just warn once.

Usage

.collectWarnings(expr, lst = FALSE)

Arguments

Value

The value of the expression or a list with the value of the expression and a list of warning messages

Author(s)

Matthew L. Fidler

Convert a factor/char to an id

Description

Convert a factor/char to an id

Usage

.convertId(a)

Arguments

Value

id factor

Author(s)

Matthew L. Fidler

Examples

.convertId("a")

This copies the rxode2 UI object so it can be modified

Description

This copies the rxode2 UI object so it can be modified

Usage

.copyUi(ui)

Arguments

Value

Copied UI object

Author(s)

Matthew L. Fidler

Expand parameters

Description

Expand parameters

Usage

.expandPars(object, params, events, control)

Arguments

Value

Expanded parameters for simulation

Author(s)

Matthew L. Fidler

Generate extraC information for rxode2 models

Description

Generate extraC information for rxode2 models

Usage

.extraC(extraC = NULL)

Arguments

Value

Nothing, called for side effects

Author(s)

Matthew L. Fidler

Get the extraCnow for compiling

Description

Get the extraCnow for compiling

Usage

.extraCnow()

Value

string of extraC information

Author(s)

Matthew L. Fidler

Get the last idLvl

Description

Get the last idLvl

Usage

.getLastIdLvl()

Value

Last idLvl

Author(s)

Matthew L. Fidler

Examples

.getLastIdLvl()

Get the internal breakdown of an evid

Description

Get the internal breakdown of an evid

Usage

.getWh(i)

Arguments

Value

named evid integer vector

Author(s)

Matthew L. Fidler

Examples

.getWh(1001)
.getWh(10401)

Handle the single error for normal or t distributions

Description

Handle the single error for normal or t distributions

Usage

.handleSingleErrTypeNormOrTFoceiBase(
  env,
  pred1,
  errNum = 1L,
  rxPredLlik = TRUE
)

Arguments

Value

A list of the lines added. The lines will contain

• rx_yj_ which is an integer that corresponds to the transformation type.

• rx_lambda_ is the transformation lambda

• rx_low_ The lower boundary of the transformation

• rx_hi_ The upper boundary of the transformation

• rx_pred_f_ The prediction function

• rx_pred_ The transformed prediction function

• rx_r_ The transformed variance

Author(s)

Matthew Fidler

This gets the append arg for the ini() piping

Description

This gets the append arg for the ini() piping

Usage

.iniGetAppendArg(f, s)

Arguments

Value

corrected ini piping argument

This is exported for creating new ini methods that have the same requirements for piping

Author(s)

Matthew L. Fidler

Reorder rows in iniDf

Description

Reorder rows in iniDf

Usage

.iniHandleAppend(expr, rxui, envir, append)

Arguments

Value

Nothing, called for side effects

Update the iniDf of a model

Description

Update the iniDf of a model

Usage

.iniHandleLine(expr, rxui, envir = parent.frame(), append = NULL)

.iniHandleFixOrUnfix(expr, rxui, envir = parent.frame(), append = NULL)

Arguments

Value

Nothing, called for side effects

Author(s)

Matthew L. Fidler

Check if a language object matches a template language object

Description

• If template == str2lang("."), it will match anything.

• If template == str2lang(".name"), it will match any name.

• If template == str2lang(".call()"), it will match any call.

Usage

.matchesLangTemplate(x, template)

Arguments

Value

TRUE if it matches, FALSE, otherwise

Examples

.matchesLangTemplate(str2lang("d/dt(foo)"), str2lang("d/dt(.name)"))
.matchesLangTemplate(str2lang("d/dt(foo)"), str2lang("d/foo(.name)"))
.matchesLangTemplate(str2lang("d/dt(foo)"), str2lang("d/."))

Internal messaging statements

Description

Internal messaging statements

Usage

.minfo(text, ..., .envir = parent.frame())

.malert(text, ..., .envir = parent.frame())

.mwarn(text, ..., .envir = parent.frame())

.msuccess(text, ..., .envir = parent.frame())

Arguments

Value

Nothing, called for side effects

Author(s)

Matthew L. Fidler

Handle model lines

Description

Handle model lines

Usage

.modelHandleModelLines(
  modelLines,
  rxui,
  modifyIni = FALSE,
  append = NULL,
  auto = getOption("rxode2.autoVarPiping", TRUE),
  cov = NULL,
  envir
)

Arguments

Value

New UI

Author(s)

Matthew L. Fidler

Assign in the rxode2 pipeline

Description

Assign in the rxode2 pipeline

Usage

.pipeRx(obj)

.pipeInits(obj)

.pipeEvents(obj)

.pipeParams(obj)

.pipeKeep(obj)

.pipeThetaMat(obj)

.pipeOmega(obj)

.pipeSigma(obj)

.pipeDfObs(obj)

.pipeDfSub(obj)

.pipeNSub(obj)

.pipeNStud(obj)

Arguments

Value

The pipeline object (invisibly)

Author(s)

Matthew L. Fidler

Returns quoted call information

Description

Returns quoted call information

Usage

.quoteCallInfoLines(callInfo, envir = parent.frame(), iniDf = NULL)

Arguments

Value

Quote call information. for name=expression, change to name<-expression in quoted call list. For expressions that are within brackets ie {}, unlist the brackets as if they were called in one single sequence.

Author(s)

Matthew L. Fidler

Get a blank, theta1, or eta1 initialization block for iniDf

Description

Get a blank, theta1, or eta1 initialization block for iniDf

Usage

.rxBlankIni(type = c("empty", "theta", "eta"))

Arguments

Value

A data.frame with the appropriate number/type of columns.

For type="empty", the data.frame will have 0 rows but all the correct types.

For type="theta", the data.frame will have 1 row with the correct types and default values. The "name" and "est" will likely need to be updated.

For type="eta", the data.frame will have 1 row with the correct types and default values for the a single eta being added. The "name" and "est" will likely need to be updated.

Author(s)

Matthew L. Fidler

Examples

.rxBlankIni("empty")

.rxBlankIni("theta")

.rxBlankIni("eta")

Return the C code of an internal function

Description

Return the C code of an internal function

Usage

.rxC(fun)

Arguments

Value

C code if found (as a string) or NULL if not found

Author(s)

Matthew Fider

Print out a table in the documentation

Description

Print out a table in the documentation

Usage

.rxDocTable(table, caption = "none")

Arguments

Value

based on the knitr context:

• output a kableExtra::kbl for latex output

• output a DT::datatable for html output

• otherwise output a knitr::kable

Author(s)

Matthew L. Fidler

Examples

.rxDocTable(rxReservedKeywords)

Get the upper boundary condition when the transformation it

Description

Get the upper boundary condition when the transformation it

Usage

.rxGetHiBoundaryPred1AndIni(env, pred1)

Arguments

Value

Upper Boundary

Author(s)

Matthew Fidler

Get the lambda value based on the pred information

Description

Get the lambda value based on the pred information

Usage

.rxGetLambdaFromPred1AndIni(env, pred1)

Arguments

Value

Lambda expression

Author(s)

Matthew Fidler

Get the lower boundary condition when the transformation requires it

Description

Get the lower boundary condition when the transformation requires it

Usage

.rxGetLowBoundaryPred1AndIni(env, pred1)

Arguments

Value

Lower Boundary

Author(s)

Matthew Fidler

Get the Observation transformation

Description

Get the Observation transformation

Usage

.rxGetPredictionDVTransform(env, pred1, yj)

Arguments

Value

The transformation expression

Author(s)

Matthew Fidler

Get the prediction name

Description

Get the prediction name

Usage

.rxGetPredictionF(env, pred1)

Arguments

Value

The prediction symbol

Author(s)

Matthew Fidler

Get the prediction transformation

Description

Get the prediction transformation

Usage

.rxGetPredictionFTransform(env, pred1, yj)

Arguments

Value

The transformation expression

Author(s)

Matthew Fidler

Get Variance for error type

Description

Get Variance for error type

Usage

.rxGetVarianceForErrorType(env, pred1)

Arguments

Value

Variance error type

Author(s)

Matthew L. Fidler

Internal function to tell if the linCmt() is the model variables

Description

Internal function to tell if the linCmt() is the model variables

Usage

.rxIsLinCmt()

Value

0 or 1

Author(s)

Matthew L. Fidler

Internal function for calculating the Jacobian

Description

Internal function for calculating the Jacobian

Usage

.rxJacobian(model, vars = TRUE)

Arguments

Value

Jacobian information

Author(s)

Matthew L. Fidler

Get the linear compartment model states

Description

Get the linear compartment model states

Usage

.rxLinCmt(obj)

Arguments

Value

compartment names

Author(s)

Matthew L. Fidler

Internal function to generate the model variables for a linCmt() model

Description

Internal function to generate the model variables for a linCmt() model

Usage

.rxLinCmtGen(lenState, vars)

Arguments

Value

Model variables of expanded linCmt model

Author(s)

Matthew L. Fidler

Get the number of linear compartments

Description

Get the number of linear compartments

Usage

.rxLinNcmt(obj)

Arguments

Value

A named integer vector with the number of linear compartments (numLin, numLinSens and depotLin)

Author(s)

Matthew L. Fidler

Internal Pruning function

Description

Internal Pruning function

Usage

.rxPrune(x, envir = parent.frame(), strAssign = list())

Arguments

Value

Pruned model code

Author(s)

Matthew L. Fidler

Return the equivalents symengine user functions from C

Description

Return the equivalents symengine user functions from C

Usage

.rxSEeqUsr()

Value

equivalent symengine user functions

Author(s)

Matthew L. Fidler

Sensitivity for model

Description

Sensitivity for model

Usage

.rxSens(model, vars, vars2)

Arguments

Value

Sensitivity

Author(s)

Matthew L. Fidler

Get the information about the rxode2 derived parameter transformation

Description

Get the information about the rxode2 derived parameter transformation

Usage

.rxTransInfo(...)

Arguments

Value

Translation information; This list contains:

• ⁠$str⁠ A named string of the parameters as seen in the underlying C/C++ code. The parameters that are NA are not used in the linear compartment model calculations.

• ⁠$ncmt⁠ the number of compartments in the model

• ⁠$trans⁠ the rxode2 translation number of the parameterization

This contains the linCmt() translation number, the number of compartments and the parameters

Author(s)

Matthew L. Fidler

Examples

.rxTransInfo(cl, v , Vp, vp2, q, q2)

.rxTransInfo(k12, k21, k13, k31, kel, v)

.rxTransInfo(k12, k21, k13, k31, kel, v, ka)

.rxTransInfo(CL, V)

rxode2 general transformation function

Description

rxode2 general transformation function

Usage

.rxTransform(
  x,
  lambda = 1,
  low = 0,
  high = 1,
  transform = c("boxCox", "yeoJohnson", "untransformed", "lnorm", "logit",
    "logit + yeoJohnson", "probit", "probit + yeoJohnson", "logit + boxCox",
    "probit + boxCox"),
  inverse = FALSE
)

Arguments

Value

transformed value

Author(s)

Matthew L. Fidler

Examples

logit(0.25)

.rxTransform(0.25, transform="logit")

expit(-1.09)

.rxTransform(-1.09, transform="logit", inverse=TRUE)

Temporarily set options then restore them while running code

Description

Temporarily set options then restore them while running code

Usage

.rxWithOptions(ops, code)

Arguments

Value

value of code

Examples

.rxWithOptions(list(digits = 21), {
  print(pi)
})

print(pi)

With one sink, then release

Description

With one sink, then release

Usage

.rxWithSink(file, code)

.rxWithSinkBoth(file, code)

Arguments

Details

.rxWithSink captures output from cat

.rxWithSinkBoth captures output from cat and message

Value

Will return the results of the code section

Author(s)

Matthew Fidler

Examples

t <- tempfile()
.rxWithSink(t, cat("message\n"))
cat("cat2\n") # now you can see the cat2
lines <- readLines(t)
unlink(t)

Temporarily set options then restore them while running code

Description

Temporarily set options then restore them while running code

Usage

.rxWithWd(wd, code)

Arguments

Value

value of code

Examples

.rxWithWd(tempdir(), {
  getwd()
})

getwd()

Get the rxode2 function pointers

Description

This function is used to get the function pointers for rxode2. This is used to allow rxode2 to have binary linkage to nlmixr2est.

Usage

.rxode2ptrs()

Value

a list of function pointers

Author(s)

Matthew L. Fidler

Examples

.rxode2ptrs()

Register a method for a suggested dependency

Description

Generally, the recommend way to register an S3 method is to use the S3Method() namespace directive (often generated automatically by the ⁠@export⁠ roxygen2 tag). However, this technique requires that the generic be in an imported package, and sometimes you want to suggest a package, and only provide a method when that package is loaded. s3_register() can be called from your package's .onLoad() to dynamically register a method only if the generic's package is loaded. (To avoid taking a dependency on vctrs for this one function, please feel free to copy and paste the function source into your own package.)

Usage

.s3register(generic, class, method = NULL)

Arguments

Details

For R 3.5.0 and later, s3_register() is also useful when demonstrating class creation in a vignette, since method lookup no longer always involves the lexical scope. For R 3.6.0 and later, you can achieve a similar effect by using "delayed method registration", i.e. placing the following in your NAMESPACE file:

if (getRversion() >= "3.6.0") {
  S3method(package::generic, class)
}

Value

nothing; called for side effects

Examples

# A typical use case is to dynamically register tibble/pillar methods
# for your class. That way you avoid creating a hard dependency on packages
# that are not essential, while still providing finer control over
# printing when they are used.

.onLoad <- function(...) {
  .s3Register("pillar::pillar_shaft", "vctrs_vctr")
  .s3Register("tibble::type_sum", "vctrs_vctr")
}

Calculate the lambdas and coefficients of the two compartment model

Description

Calculate the lambdas and coefficients of the two compartment model

Usage

.solComp2(k10, k12, k21)

Arguments

Value

List with L vector and matrices C1 and C2

Author(s)

Matthew L. Fidler based on wnl package/paper, implemented in C/C++

Examples

.solComp2(k10=0.1, k12=3, k21=1)

Calculate the lambdas and coefficients of the three compartment model

Description

Calculate the lambdas and coefficients of the three compartment model

Usage

.solComp3(k10, k12, k21, k13, k31)

Arguments

Value

List with L vector and matrices C1, C2 and C3

Author(s)

Matthew L. Fidler

Examples

.solComp3(k10=0.1, k12=3, k21=1, k13=2, k31=0.5)

Return symengineFs from user functions

Description

Return symengineFs from user functions

Usage

.symengineFs()

Value

symengineFs from user functions

Author(s)

Matthew L. Fidler

This converts NONMEM-style EVIDs to classic RxODE events

Description

This converts NONMEM-style EVIDs to classic RxODE events

Usage

.toClassicEvid(cmt = 1L, amt = 0, rate = 0, dur = 0, ii = 0, evid = 0L, ss = 0)

Arguments

Value

classic evids, excluding evids that are added (you need to add them manually) or simply use etTran. This is mostly for testing and really shouldn't be used directly.

Author(s)

Matthew L. Fidler

Examples

.toClassicEvid(cmt=10, amt=3, evid=1)
.toClassicEvid(cmt=10, amt=3, rate=2, evid=1)
.toClassicEvid(cmt=10, amt=3, rate=-1, evid=1)
.toClassicEvid(cmt=10, amt=3, rate=-2, evid=1)
.toClassicEvid(cmt=10, amt=3, dur=2, evid=1)
.toClassicEvid(cmt=304, amt=3, dur=2, evid=1)
.toClassicEvid(cmt=7, amt=0, rate=2, evid=1, ss=1)
.toClassicEvid(cmt=-10, amt=3, evid=1)
.toClassicEvid(cmt=10, amt=3, evid=5)
.toClassicEvid(cmt=6, amt=3, evid=6)
.toClassicEvid(cmt=6, amt=3, evid=7)
.toClassicEvid(evid=2)
.toClassicEvid(evid=4)

Get the function name with the current arguments as a string

Description

Get the function name with the current arguments as a string

Usage

.udfCallFunArg(fun, args)

Arguments

Value

string of the form 'fun(arg1, arg2)':

Author(s)

Matthew L. Fidler

Lock/Unlock environment for getting R user functions

Description

Lock/Unlock environment for getting R user functions

Usage

.udfEnvReset(lock = TRUE)

Arguments

Value

lock status

Author(s)

Matthew L. Fidler

Setup the UDF environment (for querying user defined funtions)

Description

Setup the UDF environment (for querying user defined funtions)

Usage

.udfEnvSet(env)

Arguments

Value

environment

Author(s)

Matthew L. Fidler

Use the udf model variable information to get the environment where the functions exists

Description

Use the udf model variable information to get the environment where the functions exists

Usage

.udfEnvSetUdf(udf)

Arguments

Value

nothing called for side effects

Author(s)

Matthew L. Fidler

See if the UI function exists in given environment.

Description

If other functions have been declared, make sure they exist too.

Usage

.udfExists(fun, nargs, envir, doList = TRUE)

Arguments

Value

logical declaring if the udf function exists in this environment

Author(s)

Matthew L. Fidler

Get the udf strings for creating model md5

Description

Get the udf strings for creating model md5

Usage

.udfMd5Info()

Value

string vector

Author(s)

Matthew L. Fidler

Handle arguments for ui functions

Description

Note this is an internal function but it is exported in case it is useful.

Usage

.uiArg(char, f, dp)

Arguments

Value

character representing the underlying rxode2 code for the argument

Author(s)

Matthew L. Fidler

Examples

.uiArg("1.0", 1.0, "1.0")

Internal function to figure out if this session supports Unicode

Description

Internal function to figure out if this session supports Unicode

Usage

.useUtf()

Value

boolean indicating if this session supports Unicode

Convert numeric vector to repeated data.frame

Description

Convert numeric vector to repeated data.frame

Usage

.vecDf(vec, n)

Arguments

Value

Data frame with repeated vec

Author(s)

Matthew Fidler

Exponential Linear Unit (ELU) Activation Function

Description

Exponential Linear Unit (ELU) Activation Function

Usage

ELU(x, alpha = 1)

Arguments

Value

A numeric vector where the ReLU function has been applied to each element of x.

Author(s)

Matthew Fidler

See Also

Other Activation Functions: GELU(), PReLU(), ReLU(), SELU(), Swish(), dELU(), dGELU(), dPReLU(), dReLU(), dSELU(), dSwish(), dlReLU(), dsoftplus(), lReLU(), softplus()

Examples

ELU(c(-1, 0, 1, 2), 2)

# Can also be used in rxode2:
x <- rxode2({
   r=SELU(time)
})

e <- et(c(-1, 0, 1, 2))

rxSolve(x, e)

GELU activation function

Description

GELU activation function

Usage

GELU(x)

Arguments

Value

numeric vector

See Also

Other Activation Functions: ELU(), PReLU(), ReLU(), SELU(), Swish(), dELU(), dGELU(), dPReLU(), dReLU(), dSELU(), dSwish(), dlReLU(), dsoftplus(), lReLU(), softplus()

Examples

GELU(c(-2, -1, 0, 1, 2))

# you can use rxode2 as well
r <- rxode2({
  r = GELU(time)
})
et <- et(c(-2, -1, 0, 1, 2))
rxSolve(r, et)

Parametric ReLU Activation Function

Description

Parametric ReLU Activation Function

Usage

PReLU(x, alpha = 1)

Arguments

Value

A numeric vector where the ReLU function has been applied to each element of x.

Author(s)

Matthew Fidler

See Also

Other Activation Functions: ELU(), GELU(), ReLU(), SELU(), Swish(), dELU(), dGELU(), dPReLU(), dReLU(), dSELU(), dSwish(), dlReLU(), dsoftplus(), lReLU(), softplus()

Examples

PReLU(c(-1, 0, 1, 2), 2)

# Can also be used in rxode2:
x <- rxode2({
   r=PReLU(time, 2)
})

e <- et(c(-1, 0, 1, 2))

rxSolve(x, e)

Rectified Linear Unit (ReLU) Activation Function

Description

This function applies the Rectified Linear Unit (ReLU) activation function to the input numeric vector. The ReLU function is defined as the positive part of its argument: f(x) = max(0, x).

Usage

ReLU(x)

Arguments

Value

A numeric vector where the ReLU function has been applied to each element of x.

Author(s)

Matthew Fidler

See Also

Other Activation Functions: ELU(), GELU(), PReLU(), SELU(), Swish(), dELU(), dGELU(), dPReLU(), dReLU(), dSELU(), dSwish(), dlReLU(), dsoftplus(), lReLU(), softplus()

Examples

ReLU(c(-1, 0, 1, 2))

# Can also be used in rxode2:
x <- rxode2({
   r=ReLU(time)
})

e <- et(c(-1, 0, 1, 2))

rxSolve(x, e)

Scaled Exponential Linear Unit (SELU) Activation Function

Description

Scaled Exponential Linear Unit (SELU) Activation Function

Usage

SELU(x)

Arguments

Value

A numeric vector where the ReLU function has been applied to each element of x.

Author(s)

Matthew Fidler

See Also

Other Activation Functions: ELU(), GELU(), PReLU(), ReLU(), Swish(), dELU(), dGELU(), dPReLU(), dReLU(), dSELU(), dSwish(), dlReLU(), dsoftplus(), lReLU(), softplus()

Examples

SELU(c(-1, 0, 1, 2))

# Can also be used in rxode2:
x <- rxode2({
   r=SELU(time)
})

e <- et(c(-1, 0, 1, 2))

rxSolve(x, e)

Switch Activation Function

Description

The switch activation function is defined as:

Usage

Swish(x)

Arguments

Details

f(x) = x \cdot \text{sigmoid}(x)

Value

A numeric vector where the ReLU function has been applied to each element of x.

Author(s)

Matthew Fidler

See Also

Other Activation Functions: ELU(), GELU(), PReLU(), ReLU(), SELU(), dELU(), dGELU(), dPReLU(), dReLU(), dSELU(), dSwish(), dlReLU(), dsoftplus(), lReLU(), softplus()

Examples

Swish(c(-1, 0, 1, 2))

# Can also be used in rxode2:
x <- rxode2({
   r<- Swish(time)
})

e <- et(c(-1, 0, 1, 2))

rxSolve(x, e)

Add dosing to eventTable

Description

This adds a dosing event to the event table. This is provided for piping syntax through magrittr. It can also be accessed by eventTable$add.dosing(...)

Usage

add.dosing(
  eventTable,
  dose,
  nbr.doses = 1L,
  dosing.interval = 24,
  dosing.to = 1L,
  rate = NULL,
  amount.units = NA_character_,
  start.time = 0,
  do.sampling = FALSE,
  time.units = NA_character_,
  ...
)

Arguments

Value

eventTable with updated dosing (note the event table will be updated anyway)

Author(s)

Matthew L. Fidler

Matthew L Fidler, Wenping Wang

References

Wang W, Hallow K, James D (2015). "A Tutorial on rxode2: Simulating Differential Equation Pharmacometric Models in R." CPT: Pharmacometrics and Systems Pharmacology, 5(1), 3-10. ISSN 2163-8306

See Also

eventTable, add.sampling, add.dosing, et, etRep, etRbind, rxode2

Examples

## Not run: 

library(rxode2)
library(units)

# Model from rxode2 tutorial
# Using a nlmixr2 style function

mod1 <-function(){
  ini({
    KA <- 2.94E-01
    CL <- 1.86E+01
    V2 <- 4.02E+01
    Q <- 1.05E+01
    V3 <- 2.97E+02
    Kin <- 1
    Kout <- 1
    EC50 <- 200
  })
 model({
    C2 <- centr/V2
    C3 <- peri/V3
    d/dt(depot) <- -KA*depot
    d/dt(centr) <- KA*depot - CL*C2 - Q*C2 + Q*C3
    d/dt(peri)  <-                    Q*C2 - Q*C3
    d/dt(eff)   <- Kin - Kout*(1-C2/(EC50+C2))*eff
 })
}

## These are making the more complex regimens of the rxode2 tutorial

## bid for 5 days
bid <- et(timeUnits="hr") |>
       et(amt=10000,ii=12,until=set_units(5, "days"))

## qd for 5 days
qd <- et(timeUnits="hr") |>
      et(amt=20000,ii=24,until=set_units(5, "days"))

## bid for 5 days followed by qd for 5 days

et <- seq(bid,qd) |>
      et(seq(0,11*24,length.out=100))

bidQd <- rxSolve(mod1, et)

plot(bidQd, C2)


## Now Infusion for 5 days followed by oral for 5 days

##  note you can dose to a named compartment instead of using the compartment number
infusion <- et(timeUnits = "hr") |>
      et(amt=10000, rate=5000, ii=24, until=set_units(5, "days"), cmt="centr")


qd <- et(timeUnits = "hr") |>
  et(amt=10000, ii=24, until=set_units(5, "days"), cmt="depot")

et <- seq(infusion,qd)

infusionQd <- rxSolve(mod1, et)

plot(infusionQd, C2)

## 2wk-on, 1wk-off

qd <- et(timeUnits = "hr") |>
      et(amt=10000, ii=24, until=set_units(2, "weeks"), cmt="depot")

et <- seq(qd, set_units(1,"weeks"), qd) |>
     add.sampling(set_units(seq(0, 5.5,by=0.005),weeks))

wkOnOff <- rxSolve(mod1, et)

plot(wkOnOff, C2)

## You can also repeat the cycle easily with the rep function

qd <-et(timeUnits = "hr") |>
     et(amt=10000, ii=24, until=set_units(2, "weeks"), cmt="depot")

et <- etRep(qd, times=4, wait=set_units(1,"weeks")) |>
      add.sampling(set_units(seq(0, 12.5,by=0.005),weeks))

repCycle4 <- rxSolve(mod1, et)

plot(repCycle4, C2)


## End(Not run)

Add sampling to eventTable

Description

This adds a dosing event to the event table. This is provided for piping syntax through magrittr. It can also be accessed by eventTable$add.sampling()

Usage

add.sampling(eventTable, time, time.units = NA)

Arguments

Value

eventTable with updated sampling. (Note the event table will be updated even if you don't reassign the eventTable)

Author(s)

Matthew L Fidler, Wenping Wang

References

Wang W, Hallow K, James D (2015). "A Tutorial on rxode2: Simulating Differential Equation Pharmacometric Models in R." CPT: Pharmacometrics and Systems Pharmacology, 5(1), 3-10. ISSN 2163-8306

See Also

eventTable, add.sampling, add.dosing, et, etRep, etRbind, rxode2

Examples

## Not run: 

library(rxode2)
library(units)

# Model from rxode2 tutorial
# Using a nlmixr2 style function

mod1 <-function(){
  ini({
    KA <- 2.94E-01
    CL <- 1.86E+01
    V2 <- 4.02E+01
    Q <- 1.05E+01
    V3 <- 2.97E+02
    Kin <- 1
    Kout <- 1
    EC50 <- 200
  })
 model({
    C2 <- centr/V2
    C3 <- peri/V3
    d/dt(depot) <- -KA*depot
    d/dt(centr) <- KA*depot - CL*C2 - Q*C2 + Q*C3
    d/dt(peri)  <-                    Q*C2 - Q*C3
    d/dt(eff)   <- Kin - Kout*(1-C2/(EC50+C2))*eff
 })
}

## These are making the more complex regimens of the rxode2 tutorial

## bid for 5 days
bid <- et(timeUnits="hr") |>
       et(amt=10000,ii=12,until=set_units(5, "days"))

## qd for 5 days
qd <- et(timeUnits="hr") |>
      et(amt=20000,ii=24,until=set_units(5, "days"))

## bid for 5 days followed by qd for 5 days

et <- seq(bid,qd) |>
      et(seq(0,11*24,length.out=100))

bidQd <- rxSolve(mod1, et)

plot(bidQd, C2)


## Now Infusion for 5 days followed by oral for 5 days

##  note you can dose to a named compartment instead of using the compartment number
infusion <- et(timeUnits = "hr") |>
      et(amt=10000, rate=5000, ii=24, until=set_units(5, "days"), cmt="centr")


qd <- et(timeUnits = "hr") |>
  et(amt=10000, ii=24, until=set_units(5, "days"), cmt="depot")

et <- seq(infusion,qd)

infusionQd <- rxSolve(mod1, et)

plot(infusionQd, C2)

## 2wk-on, 1wk-off

qd <- et(timeUnits = "hr") |>
      et(amt=10000, ii=24, until=set_units(2, "weeks"), cmt="depot")

et <- seq(qd, set_units(1,"weeks"), qd) |>
     add.sampling(set_units(seq(0, 5.5,by=0.005),weeks))

wkOnOff <- rxSolve(mod1, et)

plot(wkOnOff, C2)

## You can also repeat the cycle easily with the rep function

qd <-et(timeUnits = "hr") |>
     et(amt=10000, ii=24, until=set_units(2, "weeks"), cmt="depot")

et <- etRep(qd, times=4, wait=set_units(1,"weeks")) |>
      add.sampling(set_units(seq(0, 12.5,by=0.005),weeks))

repCycle4 <- rxSolve(mod1, et)

plot(repCycle4, C2)


## End(Not run)

Coerce object to data.frame

Description

Coerce object to data.frame

Usage

as.et(x, ...)

## Default S3 method:
as.et(x, ...)

Arguments

Value

An event table

Turn into an ini block for initialization

Description

Turn into an ini block for initialization

Usage

as.ini(x)

## S3 method for class 'character'
as.ini(x)

## S3 method for class 'data.frame'
as.ini(x)

## S3 method for class 'call'
as.ini(x)

## S3 method for class 'lotriFix'
as.ini(x)

## S3 method for class 'matrix'
as.ini(x)

## Default S3 method:
as.ini(x)

Arguments

Value

rxode2 ini expression

Author(s)

Matthew L. Fidler

Examples

ini <- quote(ini({
  tka <- log(1.57)
  tcl <- log(2.72)
  tv <- log(31.5)
  eta.ka ~ 0.6
  eta.cl ~ 0.3
  eta.v ~ 0.1
  add.sd <- 0.7
}))

as.ini(ini)

l <- quote(lotri({
  tka <- log(1.57)
  tcl <- log(2.72)
  tv <- log(31.5)
  eta.ka ~ 0.6
  eta.cl ~ 0.3
  eta.v ~ 0.1
  add.sd <- 0.7
 }))

as.ini(l)

m <- lotri({
   eta.ka ~ 0.6
   eta.cl ~ 0.3
   eta.v ~ 0.1
})

as.ini(m)

one.compartment <- function() {
  ini({
    tka <- log(1.57)
    tcl <- log(2.72)
    tv <- log(31.5)
    eta.ka ~ 0.6
    eta.cl ~ 0.3
    eta.v ~ 0.1
    add.sd <- 0.7
  })
  model({
    ka <- exp(tka + eta.ka)
    cl <- exp(tcl + eta.cl)
    v <- exp(tv + eta.v)
    d/dt(depot) = -ka * depot
    d/dt(center) = ka * depot - cl / v * center
    cp = center / v
    cp ~ add(add.sd)
  })
}

as.ini(one.compartment)

ui <- one.compartment()

as.ini(ui)

ui$iniDf

as.ini(ui$iniDf)

ini <- c("ini({",
          "tka <- log(1.57)",
          "tcl <- log(2.72)",
          "tv <- log(31.5)",
          "eta.ka ~ 0.6",
          "eta.cl ~ 0.3",
          "eta.v ~ 0.1",
          "add.sd <- 0.7",
          "})")

as.ini(ini)

ini <- paste(ini, collapse="\n")

as.ini(ini)

Turn into a model expression

Description

Turn into a model expression

Usage

as.model(x)

## S3 method for class 'character'
as.model(x)

## S3 method for class 'call'
as.model(x)

## S3 method for class 'list'
as.model(x)

## Default S3 method:
as.model(x)

Arguments

Value

model expression

Author(s)

Matthew L. Fidler

Examples

model <- quote(model({
  ka <- exp(tka + eta.ka)
  cl <- exp(tcl + eta.cl)
  v <- exp(tv + eta.v)
  d/dt(depot) = -ka * depot
  d/dt(center) = ka * depot - cl / v * center
  cp = center / v
  cp ~ add(add.sd)
}))
 
as.model(model)

one.compartment <- function() {
  ini({
    tka <- log(1.57)
    tcl <- log(2.72)
    tv <- log(31.5)
    eta.ka ~ 0.6
    eta.cl ~ 0.3
    eta.v ~ 0.1
    add.sd <- 0.7
  })
  model({
    ka <- exp(tka + eta.ka)
    cl <- exp(tcl + eta.cl)
    v <- exp(tv + eta.v)
    d/dt(depot) = -ka * depot
    d/dt(center) = ka * depot - cl / v * center
    cp = center / v
    cp ~ add(add.sd)
  })
}

as.model(one.compartment)
 
ui <- one.compartment()

as.model(ui)

model <- c("model({",
           "ka <- exp(tka + eta.ka)",
           "cl <- exp(tcl + eta.cl)",
           "v <- exp(tv + eta.v)",
           "d/dt(depot) = -ka * depot",
           "d/dt(center) = ka * depot - cl / v * center",
           "cp = center / v",
           "cp ~ add(add.sd)",
           "})")

as.model(model)

model <- paste(model, collapse="\n")

as.model(model)

As rxode2 ui

Description

As rxode2 ui

Usage

as.rxUi(x)

## S3 method for class 'rxode2'
as.rxUi(x)

## S3 method for class 'rxode2tos'
as.rxUi(x)

## S3 method for class 'rxModelVars'
as.rxUi(x)

## S3 method for class ''function''
as.rxUi(x)

## S3 method for class 'rxUi'
as.rxUi(x)

## Default S3 method:
as.rxUi(x)

Arguments

Value

rxUi object (or error if it cannot be converted)

Author(s)

Matthew L. Fidler

Examples

mod1 <- function() {
 ini({
   # central 
   KA=2.94E-01
   CL=1.86E+01
   V2=4.02E+01
   # peripheral
   Q=1.05E+01
   V3=2.97E+02
   # effects
   Kin=1
   Kout=1
   EC50=200 
 })
 model({
   C2 <- centr/V2
   C3 <- peri/V3
   d/dt(depot) <- -KA*depot
   d/dt(centr) <- KA*depot - CL*C2 - Q*C2 + Q*C3
   d/dt(peri)  <- Q*C2 - Q*C3
   eff(0) <- 1
   d/dt(eff)   <- Kin - Kout*(1-C2/(EC50+C2))*eff
 })
}

as.rxUi(mod1)

Verify that the compartment exists in a model

Description

Verify that the compartment exists in a model

Usage

assertCompartmentExists(ui, x)

testCompartmentExists(ui, x)

Arguments

Value

the value of the compartment that exists; if it is a vector returns the first item that matches

Functions

• testCompartmentExists(): Test if compartment exists

Author(s)

Matthew Fidler & Bill Denney

See Also

Other Assertions: assertCompartmentName(), assertCompartmentNew(), assertRxUi(), assertVariableExists(), assertVariableNew(), testIniDf(), testRxUnbounded()

Verify that a value is a valid nlmixr2 compartment name

Description

Verify that a value is a valid nlmixr2 compartment name

Usage

assertCompartmentName(x)

assertVariableName(x)

assertParameterValue(x)

assertExists(ui, x)

testExists(ui, x)

Arguments

Value

The value or an error

Functions

• assertVariableName(): Verify that a value is a valid nlmixr2 variable name

• assertParameterValue(): Verify that a value is a valid nlmixr2 parameter value

• assertExists(): Assert compartment/variable exists

• testExists(): Test compartment/variable exists

Author(s)

Bill Denney

See Also

Other Assertions: assertCompartmentExists(), assertCompartmentNew(), assertRxUi(), assertVariableExists(), assertVariableNew(), testIniDf(), testRxUnbounded()

Verify that a compartment would be new to the model

Description

Verify that a compartment would be new to the model

Usage

assertCompartmentNew(ui, x)

Arguments

Value

The value or an error

Author(s)

Matthew Fidler & Bill Denney

See Also

Other Assertions: assertCompartmentExists(), assertCompartmentName(), assertRxUi(), assertVariableExists(), assertVariableNew(), testIniDf(), testRxUnbounded()

Assert properties of the rxUi models

Description

Assert properties of the rxUi models

Usage

assertRxUi(ui, extra = "", .var.name = .vname(ui))

assertRxUiPrediction(ui, extra = "", .var.name = .vname(ui))

assertRxUiSingleEndpoint(ui, extra = "", .var.name = .vname(ui))

assertRxUiTransformNormal(ui, extra = "", .var.name = .vname(ui))

assertRxUiNormal(ui, extra = "", .var.name = .vname(ui))

assertRxUiMuRefOnly(ui, extra = "", .var.name = .vname(ui))

assertRxUiEstimatedResiduals(ui, extra = "", .var.name = .vname(ui))

assertRxUiPopulationOnly(ui, extra = "", .var.name = .vname(ui))

assertRxUiMixedOnly(ui, extra = "", .var.name = .vname(ui))

assertRxUiRandomOnIdOnly(ui, extra = "", .var.name = .vname(ui))

Arguments

Details

These functions have different types of assertions

• assertRxUi – Make sure this is a proper rxode2 model (if not throw error)

• assertRxUiSingleEndpoint – Make sure the rxode2 model is only a single endpoint model (if not throw error)

• assertRxUiTransformNormal – Make sure that the model residual distribution is normal or transformably normal

• assertRxUiNormal – Make sure that the model residual distribution is normal

• assertRxUiEstimatedResiduals – Make sure that the residual error parameters are estimated (not modeled).

• assertRxUiPopulationOnly – Make sure the model is the population only model (no mixed effects)

• assertRxUiMixedOnly – Make sure the model is a mixed effect model (not a population effect, only)

• assertRxUiPrediction – Make sure the model has predictions

• assertRxUiMuRefOnly – Make sure that all the parameters are mu-referenced

• assertRxUiRandomOnIdOnly – Make sure there are only random effects at the ID level

Value

the rxUi model

Author(s)

Matthew L. Fidler

See Also

Other Assertions: assertCompartmentExists(), assertCompartmentName(), assertCompartmentNew(), assertVariableExists(), assertVariableNew(), testIniDf(), testRxUnbounded()

Examples

one.cmt <- function() {
 ini({
   tka <- 0.45; label("Ka")
   tcl <- log(c(0, 2.7, 100)); label("Cl")
   tv <- 3.45; label("V")
   eta.ka ~ 0.6
   eta.cl ~ 0.3
   eta.v ~ 0.1
   add.sd <- 0.7
 })
 model({
   ka <- exp(tka + eta.ka)
   cl <- exp(tcl + eta.cl)
   v <- exp(tv + eta.v)
   linCmt() ~ add(add.sd)
 })
}

assertRxUi(one.cmt)
# assertRxUi(rnorm) # will fail

assertRxUiSingleEndpoint(one.cmt)

Assert a variable exists in the model

Description

Assert a variable exists in the model

Usage

assertVariableExists(ui, x)

testVariableExists(ui, x)

Arguments

Value

variable that matches, in the case of multiple variables, the first that matches. If nothing matches return error

Functions

• testVariableExists(): Test if variable exists

Author(s)

Matthew L. Fidler

See Also

Other Assertions: assertCompartmentExists(), assertCompartmentName(), assertCompartmentNew(), assertRxUi(), assertVariableNew(), testIniDf(), testRxUnbounded()

Assert a variable would be new to the model

Description

Assert a variable would be new to the model

Usage

assertVariableNew(ui, x)

Arguments

Value

nothing, but will error if x would not be new

Author(s)

Matthew L. Fidler

See Also

Other Assertions: assertCompartmentExists(), assertCompartmentName(), assertCompartmentNew(), assertRxUi(), assertVariableExists(), testIniDf(), testRxUnbounded()

Calculate expected confidence bands with binomial sampling distribution

Description

This is meant to perform in the same way as quantile() so it can be a drop in replacement for code using quantile() but using distributional assumptions.

Usage

binomProbs(x, ...)

## Default S3 method:
binomProbs(
  x,
  probs = c(0.025, 0.05, 0.5, 0.95, 0.975),
  na.rm = FALSE,
  names = TRUE,
  onlyProbs = TRUE,
  n = 0L,
  m = 0L,
  pred = FALSE,
  piMethod = c("lim"),
  M = 5e+05,
  tol = .Machine$double.eps^0.25,
  ciMethod = c("wilson", "wilsonCorrect", "agrestiCoull", "wald", "wc", "ac"),
  ...
)

Arguments

Details

It is used for confidence intervals with rxode2 solved objects using confint(mean="binom")

Value

By default the return has the probabilities as names (if named) with the points where the expected distribution are located given the sampling mean and standard deviation. If onlyProbs=FALSE then it would prepend mean, variance, standard deviation, minimum, maximum and number of non-NA observations.

Author(s)

Matthew L. Fidler

References

• Newcombe, R. G. (1998). "Two-sided confidence intervals for the single proportion: comparison of seven methods". Statistics in Medicine. 17 (8): 857–872. doi:10.1002/(SICI)1097-0258(19980430)17:8<857::AID-SIM777>3.0.CO;2-E. PMID 9595616.

• Hezhi Lu, Hua Jin, A new prediction interval for binomial random variable based on inferential models, Journal of Statistical Planning and Inference, Volume 205, 2020, Pages 156-174, ISSN 0378-3758, https://doi.org/10.1016/j.jspi.2019.07.001.

Examples

x<- rbinom(7001, p=0.375, size=1)
binomProbs(x)

# you can also use the prediction interval

binomProbs(x, pred=TRUE)


# Can get some extra statistics if you request onlyProbs=FALSE
binomProbs(x, onlyProbs=FALSE)

x[2] <- NA_real_

binomProbs(x, onlyProbs=FALSE)

binomProbs(x, na.rm=TRUE)

boxCox/yeoJohnson and inverse boxCox/yeoJohnson functions

Description

boxCox/yeoJohnson and inverse boxCox/yeoJohnson functions

Usage

boxCox(x, lambda = 1)

boxCoxInv(x, lambda = 1)

yeoJohnson(x, lambda = 1)

yeoJohnsonInv(x, lambda = 1)

Arguments

Value

values from boxCox and boxCoxInv

Examples

boxCox(10, 0.5)

boxCoxInv(4.32, 0.5)

yeoJohnson(10, 0.5)

yeoJohnsonInv(4.32, 0.5)

Return the rxode2 coefficients

Description

This returns the parameters , state variables

Usage

## S3 method for class 'rxode2'
coef(object, ...)

Arguments

Value

a rxCoef object with the following

• params is a list of strings for parameters for the rxode2 object

• state is a list of strings for the names of each state in the rxode2 object.

• ini is the model specified default values for the parameters.

• rxode2 is the referring rxode2 object

Author(s)

Matthew L.Fidler

Sample a covariance Matrix from the Posterior Inverse Wishart distribution.

Description

Note this Inverse wishart rescaled to match the original scale of the covariance matrix.

Usage

cvPost(
  nu,
  omega,
  n = 1L,
  omegaIsChol = FALSE,
  returnChol = FALSE,
  type = c("invWishart", "lkj", "separation"),
  diagXformType = c("log", "identity", "variance", "nlmixrSqrt", "nlmixrLog",
    "nlmixrIdentity")
)

Arguments

Details

If your covariance matrix is a 1x1 matrix, this uses an scaled inverse chi-squared which is equivalent to the Inverse Wishart distribution in the uni-directional case.

In general, the separation strategy is preferred for diagonal matrices. If the dimension of the matrix is below 10, lkj is numerically faster than separation method. However, the lkj method has densities too close to zero (XXXX) when the dimension is above 10. In that case, though computationally more expensive separation method performs better.

For matrices with modeled covariances, the easiest method to use is the inverse Wishart which allows the simulation of correlation matrices (XXXX). This method is more well suited for well behaved matrices, that is the variance components are not too low or too high. When modeling nonlinear mixed effects modeling matrices with too high or low variances are considered sub-optimal in describing a system. With these rules in mind, it is reasonable to use the inverse Wishart.

Value

a matrix (n=1) or a list of matrices (n > 1)

Author(s)

Matthew L.Fidler & Wenping Wang

References

Alvarez I, Niemi J and Simpson M. (2014) Bayesian Inference for a Covariance Matrix. Conference on Applied Statistics in Agriculture.

Wang1 Z, Wu Y, and Chu H. (2018) On Equivalence of the LKJ distribution and the restricted Wishart distribution. <doi:10.48550/arXiv.1809.047463

Examples

## Sample a single covariance.
draw1 <- cvPost(3, matrix(c(1, .3, .3, 1), 2, 2))

## Sample 3 covariances
set.seed(42)
draw3 <- cvPost(3, matrix(c(1, .3, .3, 1), 2, 2), n = 3)

## Sample 3 covariances, but return the cholesky decomposition
set.seed(42)
draw3c <- cvPost(3, matrix(c(1, .3, .3, 1), 2, 2), n = 3, returnChol = TRUE)

## Sample 3 covariances with lognormal standard deviations via LKJ
## correlation sample
cvPost(3, sapply(1:3, function(...) {
  rnorm(10)
}), type = "lkj")

## or return cholesky decomposition
cvPost(3, sapply(1:3, function(...) {
  rnorm(10)
}),
type = "lkj",
returnChol = TRUE
)

## Sample 3 covariances with lognormal standard deviations via separation
## strategy using inverse Wishart correlation sample
cvPost(3, sapply(1:3, function(...) {
  rnorm(10)
}), type = "separation")

## or returning the cholesky decomposition
cvPost(3, sapply(1:3, function(...) {
  rnorm(10)
}),
type = "separation",
returnChol = TRUE
)

Derivatives of the Exponential Linear Unit (ELU) Activation Function

Description

Derivatives of the Exponential Linear Unit (ELU) Activation Function

Usage

dELU(x, alpha = 1)

d2ELU(x, alpha = 1)

d2aELU(x, alpha = 1)

dELUa(x, alpha = 1)

d2ELUa(x, alpha = 1)

Arguments

Value

A numeric vector where the derivative(s) of the ELU function has been applied to each element of x.

Author(s)

Matthew L. Fidler

See Also

Other Activation Functions: ELU(), GELU(), PReLU(), ReLU(), SELU(), Swish(), dGELU(), dPReLU(), dReLU(), dSELU(), dSwish(), dlReLU(), dsoftplus(), lReLU(), softplus()

Examples

dELU(c(-1, 0, 1, 2), 2)
d2ELU(c(-1, 0, 1, 2), 2)
d2aELU(c(-1, 0, 1, 2), 2)
dELUa(c(-1, 0, 1, 2), 2)
d2ELUa(c(-1, 0, 1, 2), 2)

# Can also be used in rxode2:
r <- rxode2({
  r1=dELU(time, 2)
  r2=d2ELU(time, 2)
  r2a=d2aELU(time, 2)
  ra=dELUa(time, 2)
  r2a=d2ELUa(time, 2)
})

e <- et(c(-1, 0, 1, 2))
rxSolve(r, e)

Derivatives of GELU

Description

Derivatives of GELU

Usage

dGELU(x)

d2GELU(x)

d3GELU(x)

d4GELU(x)

Arguments

Value

numeric vector

See Also

Other Activation Functions: ELU(), GELU(), PReLU(), ReLU(), SELU(), Swish(), dELU(), dPReLU(), dReLU(), dSELU(), dSwish(), dlReLU(), dsoftplus(), lReLU(), softplus()

Examples

dGELU(c(-2, -1, 0, 1, 2))
d2GELU(c(-2, -1, 0, 1, 2))
d3GELU(c(-2, -1, 0, 1, 2))
d4GELU(c(-2, -1, 0, 1, 2))
# you can use rxode2 as well
r <- rxode2({
   r1 <- dGELU(time)
   r2 <- d2GELU(time)
   r3 <- d3GELU(time)
   r4 <- d4GELU(time)
})
et <- et(c(-2, -1, 0, 1, 2))
rxSolve(r, et)

Derivatives Parametric ReLU Activation Function

Description

Derivatives Parametric ReLU Activation Function

Usage

dPReLU(x, alpha = 1)

dPReLUa(x, alpha = 1)

dPReLUa1(x, alpha = 1)

Arguments

Value

A numeric vector where the derivative(s) of the ELU function has been applied to each element of x.

Author(s)

Matthew L. Fidler

See Also

Other Activation Functions: ELU(), GELU(), PReLU(), ReLU(), SELU(), Swish(), dELU(), dGELU(), dReLU(), dSELU(), dSwish(), dlReLU(), dsoftplus(), lReLU(), softplus()

Examples

dPReLU(c(-1, 0, 1, 2), 2)
dPReLUa(c(-1, 0, 1, 2), 2)
dPReLUa1(c(-1, 0, 1, 2), 2)


# Can also be used in rxode2:
r <- rxode2({
  r1=dPReLU(time, 2)
  r2a=dPReLUa(time, 2)
  ra=dPReLUa1(time, 2)
})

e <- et(c(-1, 0, 1, 2))
rxSolve(r, e)

Derivative of the Rectified Linear Unit (ReLU) Activation Function

Description

This function applies the derivative of the Rectified Linear Unit (ReLU) activation function to the input numeric vector.

Usage

dReLU(x)

Arguments

Value

A numeric vector where the derivative of the ReLU function

See Also

Other Activation Functions: ELU(), GELU(), PReLU(), ReLU(), SELU(), Swish(), dELU(), dGELU(), dPReLU(), dSELU(), dSwish(), dlReLU(), dsoftplus(), lReLU(), softplus()

Examples

dReLU(c(-1, 0, 1, 2))

# Can also be used in rxode2:
x <- rxode2({
   r=dReLU(time)
})

e <- et(c(-1, 0, 1, 2))

rxSolve(x, e)

Derivative of the Scaled Exponential Linear Unit (SELU) Activation Function

Description

Derivative of the Scaled Exponential Linear Unit (SELU) Activation Function

Usage

dSELU(x)

Arguments

Value

A numeric vector where the derivative of the SELU function has been applied to each element of x.

Author(s)

Matthew Fidler

See Also

Other Activation Functions: ELU(), GELU(), PReLU(), ReLU(), SELU(), Swish(), dELU(), dGELU(), dPReLU(), dReLU(), dSwish(), dlReLU(), dsoftplus(), lReLU(), softplus()

Examples

dSELU(c(-1, 0, 1, 2))
# Can also be used in rxode2:
x <- rxode2({
  r=dSELU(time)
})
e <- et(c(-1, 0, 1, 2))
rxSolve(x, e)

Derivative of the Swish Activation Function

Description

Derivative of the Swish Activation Function

Usage

dSwish(x)

Arguments

Value

A numeric vector where the derivative of the SELU function has been applied to each element of x.

Author(s)

Matthew Fidler

See Also

Other Activation Functions: ELU(), GELU(), PReLU(), ReLU(), SELU(), Swish(), dELU(), dGELU(), dPReLU(), dReLU(), dSELU(), dlReLU(), dsoftplus(), lReLU(), softplus()

Examples

dSwish(c(-1, 0, 1, 2))

# Can also be used in rxode2:
x <- rxode2({
  r <- dSwish(time)
})
e <- et(c(-1, 0, 1, 2))
rxSolve(x, e)

This uses simulations to match the rse

Description

This uses simulations to match the rse

Usage

dfWishart(omega, n, rse, upper, totN = 1000, diag = TRUE, seed = 1234)

Arguments

Value

output from uniroot() to find the right estimate

Author(s)

Matthew L. Fidler

Examples

dfWishart(lotri::lotri(a+b~c(1, 0.5, 1)), 100)

Derivative of Leaky ReLU activation function

Description

Derivative of Leaky ReLU activation function

Usage

dlReLU(x)

Arguments

Value

numeric vector

See Also

Other Activation Functions: ELU(), GELU(), PReLU(), ReLU(), SELU(), Swish(), dELU(), dGELU(), dPReLU(), dReLU(), dSELU(), dSwish(), dsoftplus(), lReLU(), softplus()

Examples

dlReLU(c(-1, 0, 1))

# Can use in rxode2 as well

r <- rxode2({r <- dlReLU(time)})
e <- et(c(-1, 0, 1))
rxSolve(r, e)

Default Softplus Activation Function

Description

Default Softplus Activation Function

Usage

dsoftplus(x)

d2softplus(x)

d3softplus(x)

d4softplus(x)

Arguments

Value

numeric vector

Author(s)

Matthew L. Fidler

See Also

Other Activation Functions: ELU(), GELU(), PReLU(), ReLU(), SELU(), Swish(), dELU(), dGELU(), dPReLU(), dReLU(), dSELU(), dSwish(), dlReLU(), lReLU(), softplus()

Examples

dsoftplus(c(-1, 0, 1, 2))

# You can use rxode2 too:

r <- rxode2({
 s1 <- dsoftplus(time)
})

e <- et(c(-1, 0, 1, 2))

rxSolve(r, e)

Error function

Description

Error function

Usage

erf(x)

Arguments

Value

erf of x

Author(s)

Matthew L. Fidler

Examples

erf(1.0)

Event Table Function

Description

Event Table Function

Usage

et(x, ..., envir = parent.frame())

## S3 method for class 'rxode2'
et(x, ..., envir = parent.frame())

## S3 method for class ''function''
et(x, ..., envir = parent.frame())

## S3 method for class 'rxUi'
et(x, ..., envir = parent.frame())

## S3 method for class 'rxSolve'
et(x, ..., envir = parent.frame())

## S3 method for class 'rxParams'
et(x, ..., envir = parent.frame())

## Default S3 method:
et(
  x,
  ...,
  time,
  amt,
  evid,
  cmt,
  ii,
  addl,
  ss,
  rate,
  dur,
  until,
  id,
  amountUnits,
  timeUnits,
  addSampling,
  envir = parent.frame(),
  by = NULL,
  length.out = NULL
)

Arguments

Can also specify `cmt` as `dosing.to`,
`dose.to`, `doseTo`, `dosingTo`, and
`state`.

Value

A new event table

Author(s)

Matthew L Fidler, Wenping Wang

References

Wang W, Hallow K, James D (2015). "A Tutorial on rxode2: Simulating Differential Equation Pharmacometric Models in R." CPT: Pharmacometrics and Systems Pharmacology, 5(1), 3-10. ISSN 2163-8306

See Also

eventTable, add.sampling, add.dosing, et, etRep, etRbind, rxode2

Examples

## Not run: 

library(rxode2)
library(units)

# Model from rxode2 tutorial
# Using a nlmixr2 style function

mod1 <-function(){
  ini({
    KA <- 2.94E-01
    CL <- 1.86E+01
    V2 <- 4.02E+01
    Q <- 1.05E+01
    V3 <- 2.97E+02
    Kin <- 1
    Kout <- 1
    EC50 <- 200
  })
 model({
    C2 <- centr/V2
    C3 <- peri/V3
    d/dt(depot) <- -KA*depot
    d/dt(centr) <- KA*depot - CL*C2 - Q*C2 + Q*C3
    d/dt(peri)  <-                    Q*C2 - Q*C3
    d/dt(eff)   <- Kin - Kout*(1-C2/(EC50+C2))*eff
 })
}

## These are making the more complex regimens of the rxode2 tutorial

## bid for 5 days
bid <- et(timeUnits="hr") |>
       et(amt=10000,ii=12,until=set_units(5, "days"))

## qd for 5 days
qd <- et(timeUnits="hr") |>
      et(amt=20000,ii=24,until=set_units(5, "days"))

## bid for 5 days followed by qd for 5 days

et <- seq(bid,qd) |>
      et(seq(0,11*24,length.out=100))

bidQd <- rxSolve(mod1, et)

plot(bidQd, C2)


## Now Infusion for 5 days followed by oral for 5 days

##  note you can dose to a named compartment instead of using the compartment number
infusion <- et(timeUnits = "hr") |>
      et(amt=10000, rate=5000, ii=24, until=set_units(5, "days"), cmt="centr")


qd <- et(timeUnits = "hr") |>
  et(amt=10000, ii=24, until=set_units(5, "days"), cmt="depot")

et <- seq(infusion,qd)

infusionQd <- rxSolve(mod1, et)

plot(infusionQd, C2)

## 2wk-on, 1wk-off

qd <- et(timeUnits = "hr") |>
      et(amt=10000, ii=24, until=set_units(2, "weeks"), cmt="depot")

et <- seq(qd, set_units(1,"weeks"), qd) |>
     add.sampling(set_units(seq(0, 5.5,by=0.005),weeks))

wkOnOff <- rxSolve(mod1, et)

plot(wkOnOff, C2)

## You can also repeat the cycle easily with the rep function

qd <-et(timeUnits = "hr") |>
     et(amt=10000, ii=24, until=set_units(2, "weeks"), cmt="depot")

et <- etRep(qd, times=4, wait=set_units(1,"weeks")) |>
      add.sampling(set_units(seq(0, 12.5,by=0.005),weeks))

repCycle4 <- rxSolve(mod1, et)

plot(repCycle4, C2)


## End(Not run)

Expand additional doses

Description

Expand additional doses

Usage

etExpand(et)

Arguments

Value

New event table with addl doses expanded

Author(s)

Matthew Fidler

Examples

ev <- et(amt = 3, ii = 24, until = 240)
print(ev)
etExpand(ev) # expands event table, but doesn't modify it

print(ev)

ev$expand() ## Expands the current event table and saves it in ev

Combining event tables

Description

Combining event tables

Usage

etRbind(
  ...,
  samples = c("use", "clear"),
  waitII = c("smart", "+ii"),
  id = c("merge", "unique")
)

## S3 method for class 'rxEt'
rbind(..., deparse.level = 1)

Arguments

Value

An event table

Author(s)

Matthew L Fidler

Matthew L Fidler, Wenping Wang

References

Wang W, Hallow K, James D (2015). "A Tutorial on rxode2: Simulating Differential Equation Pharmacometric Models in R." CPT: Pharmacometrics and Systems Pharmacology, 5(1), 3-10. ISSN 2163-8306

See Also

eventTable, add.sampling, add.dosing, et, etRep, etRbind, rxode2

Examples

## Not run: 

library(rxode2)
library(units)

# Model from rxode2 tutorial
# Using a nlmixr2 style function

mod1 <-function(){
  ini({
    KA <- 2.94E-01
    CL <- 1.86E+01
    V2 <- 4.02E+01
    Q <- 1.05E+01
    V3 <- 2.97E+02
    Kin <- 1
    Kout <- 1
    EC50 <- 200
  })
 model({
    C2 <- centr/V2
    C3 <- peri/V3
    d/dt(depot) <- -KA*depot
    d/dt(centr) <- KA*depot - CL*C2 - Q*C2 + Q*C3
    d/dt(peri)  <-                    Q*C2 - Q*C3
    d/dt(eff)   <- Kin - Kout*(1-C2/(EC50+C2))*eff
 })
}

## These are making the more complex regimens of the rxode2 tutorial

## bid for 5 days
bid <- et(timeUnits="hr") |>
       et(amt=10000,ii=12,until=set_units(5, "days"))

## qd for 5 days
qd <- et(timeUnits="hr") |>
      et(amt=20000,ii=24,until=set_units(5, "days"))

## bid for 5 days followed by qd for 5 days

et <- seq(bid,qd) |>
      et(seq(0,11*24,length.out=100))

bidQd <- rxSolve(mod1, et)

plot(bidQd, C2)


## Now Infusion for 5 days followed by oral for 5 days

##  note you can dose to a named compartment instead of using the compartment number
infusion <- et(timeUnits = "hr") |>
      et(amt=10000, rate=5000, ii=24, until=set_units(5, "days"), cmt="centr")


qd <- et(timeUnits = "hr") |>
  et(amt=10000, ii=24, until=set_units(5, "days"), cmt="depot")

et <- seq(infusion,qd)

infusionQd <- rxSolve(mod1, et)

plot(infusionQd, C2)

## 2wk-on, 1wk-off

qd <- et(timeUnits = "hr") |>
      et(amt=10000, ii=24, until=set_units(2, "weeks"), cmt="depot")

et <- seq(qd, set_units(1,"weeks"), qd) |>
     add.sampling(set_units(seq(0, 5.5,by=0.005),weeks))

wkOnOff <- rxSolve(mod1, et)

plot(wkOnOff, C2)

## You can also repeat the cycle easily with the rep function

qd <-et(timeUnits = "hr") |>
     et(amt=10000, ii=24, until=set_units(2, "weeks"), cmt="depot")

et <- etRep(qd, times=4, wait=set_units(1,"weeks")) |>
      add.sampling(set_units(seq(0, 12.5,by=0.005),weeks))

repCycle4 <- rxSolve(mod1, et)

plot(repCycle4, C2)


## End(Not run)

Repeat an rxode2 event table

Description

Repeat an rxode2 event table

Usage

etRep(
  x,
  times = 1,
  length.out = NA,
  each = NA,
  n = NULL,
  wait = 0,
  id = integer(0),
  samples = c("clear", "use"),
  waitII = c("smart", "+ii"),
  ii = 24
)

## S3 method for class 'rxEt'
rep(x, ...)

Arguments

Value

An event table

Author(s)

Matthew L Fidler, Wenping Wang

References

Wang W, Hallow K, James D (2015). "A Tutorial on rxode2: Simulating Differential Equation Pharmacometric Models in R." CPT: Pharmacometrics and Systems Pharmacology, 5(1), 3-10. ISSN 2163-8306

See Also

eventTable, add.sampling, add.dosing, et, etRep, etRbind, rxode2

Examples

## Not run: 

library(rxode2)
library(units)

# Model from rxode2 tutorial
# Using a nlmixr2 style function

mod1 <-function(){
  ini({
    KA <- 2.94E-01
    CL <- 1.86E+01
    V2 <- 4.02E+01
    Q <- 1.05E+01
    V3 <- 2.97E+02
    Kin <- 1
    Kout <- 1
    EC50 <- 200
  })
 model({
    C2 <- centr/V2
    C3 <- peri/V3
    d/dt(depot) <- -KA*depot
    d/dt(centr) <- KA*depot - CL*C2 - Q*C2 + Q*C3
    d/dt(peri)  <-                    Q*C2 - Q*C3
    d/dt(eff)   <- Kin - Kout*(1-C2/(EC50+C2))*eff
 })
}

## These are making the more complex regimens of the rxode2 tutorial

## bid for 5 days
bid <- et(timeUnits="hr") |>
       et(amt=10000,ii=12,until=set_units(5, "days"))

## qd for 5 days
qd <- et(timeUnits="hr") |>
      et(amt=20000,ii=24,until=set_units(5, "days"))

## bid for 5 days followed by qd for 5 days

et <- seq(bid,qd) |>
      et(seq(0,11*24,length.out=100))

bidQd <- rxSolve(mod1, et)

plot(bidQd, C2)


## Now Infusion for 5 days followed by oral for 5 days

##  note you can dose to a named compartment instead of using the compartment number
infusion <- et(timeUnits = "hr") |>
      et(amt=10000, rate=5000, ii=24, until=set_units(5, "days"), cmt="centr")


qd <- et(timeUnits = "hr") |>
  et(amt=10000, ii=24, until=set_units(5, "days"), cmt="depot")

et <- seq(infusion,qd)

infusionQd <- rxSolve(mod1, et)

plot(infusionQd, C2)

## 2wk-on, 1wk-off

qd <- et(timeUnits = "hr") |>
      et(amt=10000, ii=24, until=set_units(2, "weeks"), cmt="depot")

et <- seq(qd, set_units(1,"weeks"), qd) |>
     add.sampling(set_units(seq(0, 5.5,by=0.005),weeks))

wkOnOff <- rxSolve(mod1, et)

plot(wkOnOff, C2)

## You can also repeat the cycle easily with the rep function

qd <-et(timeUnits = "hr") |>
     et(amt=10000, ii=24, until=set_units(2, "weeks"), cmt="depot")

et <- etRep(qd, times=4, wait=set_units(1,"weeks")) |>
      add.sampling(set_units(seq(0, 12.5,by=0.005),weeks))

repCycle4 <- rxSolve(mod1, et)

plot(repCycle4, C2)


## End(Not run)

Sequence of event tables

Description

This combines a sequence of event tables.

Usage

etSeq(..., samples = c("clear", "use"), waitII = c("smart", "+ii"), ii = 24)

## S3 method for class 'rxEt'
seq(...)

Arguments

Details

This sequences all the event tables in added in the argument list .... By default when combining the event tables the offset is at least by the last inter-dose interval in the prior event table (or ii). If you separate any of the event tables by a number, the event tables will be separated at least the wait time defined by that number or the last inter-dose interval.

Value

An event table

Author(s)

Matthew L Fidler, Wenping Wang

References

Wang W, Hallow K, James D (2015). "A Tutorial on rxode2: Simulating Differential Equation Pharmacometric Models in R." CPT: Pharmacometrics and Systems Pharmacology, 5(1), 3-10. ISSN 2163-8306

See Also

eventTable, add.sampling, add.dosing, et, etRep, etRbind, rxode2

Examples

## Not run: 

library(rxode2)
library(units)

# Model from rxode2 tutorial
# Using a nlmixr2 style function

mod1 <-function(){
  ini({
    KA <- 2.94E-01
    CL <- 1.86E+01
    V2 <- 4.02E+01
    Q <- 1.05E+01
    V3 <- 2.97E+02
    Kin <- 1
    Kout <- 1
    EC50 <- 200
  })
 model({
    C2 <- centr/V2
    C3 <- peri/V3
    d/dt(depot) <- -KA*depot
    d/dt(centr) <- KA*depot - CL*C2 - Q*C2 + Q*C3
    d/dt(peri)  <-                    Q*C2 - Q*C3
    d/dt(eff)   <- Kin - Kout*(1-C2/(EC50+C2))*eff
 })
}

## These are making the more complex regimens of the rxode2 tutorial

## bid for 5 days
bid <- et(timeUnits="hr") |>
       et(amt=10000,ii=12,until=set_units(5, "days"))

## qd for 5 days
qd <- et(timeUnits="hr") |>
      et(amt=20000,ii=24,until=set_units(5, "days"))

## bid for 5 days followed by qd for 5 days

et <- seq(bid,qd) |>
      et(seq(0,11*24,length.out=100))

bidQd <- rxSolve(mod1, et)

plot(bidQd, C2)


## Now Infusion for 5 days followed by oral for 5 days

##  note you can dose to a named compartment instead of using the compartment number
infusion <- et(timeUnits = "hr") |>
      et(amt=10000, rate=5000, ii=24, until=set_units(5, "days"), cmt="centr")


qd <- et(timeUnits = "hr") |>
  et(amt=10000, ii=24, until=set_units(5, "days"), cmt="depot")

et <- seq(infusion,qd)

infusionQd <- rxSolve(mod1, et)

plot(infusionQd, C2)

## 2wk-on, 1wk-off

qd <- et(timeUnits = "hr") |>
      et(amt=10000, ii=24, until=set_units(2, "weeks"), cmt="depot")

et <- seq(qd, set_units(1,"weeks"), qd) |>
     add.sampling(set_units(seq(0, 5.5,by=0.005),weeks))

wkOnOff <- rxSolve(mod1, et)

plot(wkOnOff, C2)

## You can also repeat the cycle easily with the rep function

qd <-et(timeUnits = "hr") |>
     et(amt=10000, ii=24, until=set_units(2, "weeks"), cmt="depot")

et <- etRep(qd, times=4, wait=set_units(1,"weeks")) |>
      add.sampling(set_units(seq(0, 12.5,by=0.005),weeks))

repCycle4 <- rxSolve(mod1, et)

plot(repCycle4, C2)


## End(Not run)

Event translation for rxode2

Description

Event translation for rxode2

Usage

etTrans(
  inData,
  obj,
  addCmt = FALSE,
  dropUnits = FALSE,
  allTimeVar = FALSE,
  keepDosingOnly = FALSE,
  combineDvid = NULL,
  keep = character(0),
  addlKeepsCov = FALSE,
  addlDropSs = TRUE,
  ssAtDoseTime = TRUE,
  iCov = NULL
)

Arguments

Value

Object for solving in rxode2

Create an event table object

Description

Initializes an object of class ‘EventTable’ with methods for adding and querying dosing and observation records

Usage

eventTable(amount.units = NA, time.units = NA)

Arguments

Value

A modified data.frame with the following accessible functions:

• get.EventTable() returns the current event table

• add.dosing() adds dosing records to the event table.

• get.dosing() returns a data.frame of dosing records.

• clear.dosing() clears or deletes all dosing from event table

• 'add.sampling() adds sampling time observation records to the event table.

• get.sampling()returns a data.frame of sampled observation records.

• clear.sampling() removes all sampling from event table.

• get.obs.rec() returns a logical vector indicating whether each event record represents an observation or not.

• get.nobs() returns the number of observation (not dosing) records.

• get.units() returns a two-element character vector with the dosing and time units, respectively

• copy() makes a copy of the current event table. To create a copy of an event table object use qd2 <- qd$copy()

• expand() Expands the event table for multi-subject solving. This is done by qd$expand(400) for a 400 subject data expansion

Author(s)

Matthew Fidler, Melissa Hallow and Wenping Wang

See Also

et()

Examples

# create dosing and observation (sampling) events
# QD 50mg dosing, 5 days followed by 25mg 5 days
#
qd <- eventTable(amount.units = "mg", time.units = "days")
#
qd$add.dosing(dose = 50, nbr.doses = 5, dosing.interval = 1, do.sampling = FALSE)
#
# sample the system's drug amounts hourly the first day, then every 12 hours
# for the next 4 days
qd$add.sampling(seq(from = 0, to = 1, by = 1 / 24))
qd$add.sampling(seq(from = 1, to = 5, by = 12 / 24))
#
# print(qd$get.dosing())     # table of dosing records
print(qd$get.nobs()) # number of observation (not dosing) records
#
# BID dosing, 5 days
bid <- eventTable("mg", "days") # only dosing
bid$add.dosing(
  dose = 10000, nbr.doses = 2 * 5,
  dosing.interval = 12, do.sampling = FALSE
)
#
# Use the copy() method to create a copy (clone) of an existing
# event table (simple assignments just create a new reference to
# the same event table object (closure)).
#
bid.ext <- bid$copy() # three-day extension for a 2nd cohort
bid.ext$add.dosing(
  dose = 5000, nbr.doses = 2 * 3,
  start.time = 120, dosing.interval = 12, do.sampling = FALSE
)

# You can also use the Piping operator to create a table

qd2 <- eventTable(amount.units = "mg", time.units = "days") %>%
  add.dosing(dose = 50, nbr.doses = 5, dosing.interval = 1, do.sampling = FALSE) %>%
  add.sampling(seq(from = 0, to = 1, by = 1 / 24)) %>%
  add.sampling(seq(from = 1, to = 5, by = 12 / 24))
# print(qd2$get.dosing())     # table of dosing records
print(qd2$get.nobs()) # number of observation (not dosing) records

# Note that piping with %>% will update the original table.

qd3 <- qd2 %>% add.sampling(seq(from = 5, to = 10, by = 6 / 24))
print(qd2$get.nobs())
print(qd3$get.nobs())

Find the assignments in R expression

Description

Find the assignments in R expression

Usage

findLhs(x)

Arguments

Value

list of assigned parameters

Author(s)

Hadley Wickham and Matthew L. Fidler

Force using base order for rxode2 radix sorting

Description

Force using base order for rxode2 radix sorting

Usage

forderForceBase(forceBase = FALSE)

Arguments

Value

value of forceBase (can change if data.table is not available)

Examples

forderForceBase(TRUE) # Use base `order` for rxode2 sorts
forderForceBase(FALSE) # Use `data.table` for rxode2 sorts

Gammap: normalized lower incomplete gamma function

Description

This is the gamma_p from the boost library

Usage

gammap(a, z)

Arguments

Details

The gamma p function is given by:

gammap = lowergamma(a, z)/gamma(a)

Value

gammap results

Author(s)

Matthew L. Fidler

Examples

gammap(1, 3)
gammap(1:3, 3)
gammap(1, 1:3)

gammapDer: derivative of gammap

Description

This is the gamma_p_derivative from the boost library

Usage

gammapDer(a, z)

Arguments

Value

lowergamma results

Author(s)

Matthew L. Fidler

Examples

gammapDer(1:3, 3)

gammapDer(1, 1:3)

gammapInv and gammapInva: Inverses of normalized gammap function

Description

gammapInv and gammapInva: Inverses of normalized gammap function

Usage

gammapInv(a, p)

gammapInva(x, p)

Arguments

Details

With the equation:

p = gammap(a, x)

The 'gammapInv' function returns a value 'x' that satisfies the equation above

The 'gammapInva' function returns a value 'q' that satisfies the equation above

NOTE: gammapInva is slow

Value

inverse gammap results

Author(s)

Matthew L. Fidler

Examples

gammapInv(1:3, 0.5)

gammapInv(1, 1:3 / 3.1)

gammapInv(1:3, 1:3 / 3.1)

gammapInva(1:3, 1:3 / 3.1)

Gammaq: normalized upper incomplete gamma function

Description

This is the gamma_q from the boost library

Usage

gammaq(a, z)

Arguments

Details

The gamma q function is given by:

gammaq = uppergamma(a, z)/gamma(a)

Value

gammaq results

Author(s)

Matthew L. Fidler

Examples

gammaq(1, 3)
gammaq(1:3, 3)
gammaq(1, 1:3)

gammaqInv and gammaqInva: Inverses of normalized gammaq function

Description

gammaqInv and gammaqInva: Inverses of normalized gammaq function

Usage

gammaqInv(a, q)

gammaqInva(x, q)

Arguments

Details

With the equation:

q = gammaq(a, x)

The 'gammaqInv' function returns a value 'x' that satisfies the equation above

The 'gammaqInva' function returns a value 'a' that satisfies the equation above

NOTE: gammaqInva is slow

Value

inverse gammaq results

Author(s)

Matthew L. Fidler

Examples

gammaqInv(1:3, 0.5)

gammaqInv(1, 1:3 / 3)

gammaqInv(1:3, 1:3 / 3.1)

gammaqInva(1:3, 1:3 / 3.1)

Generate an example (template) of a dosing regimen shiny app

Description

Create a complete shiny application for exploring dosing regimens given a (hardcoded) PK/PD model.

Usage

genShinyApp.template(
  appDir = "shinyExample",
  verbose = TRUE,
  ODE.config = list(ode = "model", params = c(KA = 0.294), inits = c(eff = 1), method =
    "lsoda", atol = 1e-08, rtol = 1e-06)
)

write.template.server(appDir)

write.template.ui(appDir, statevars)

Arguments

    C2 = centr/V2;
    C3 = peri/V3;
    d/dt(depot) =-KA*depot;
    d/dt(centr) = KA*depot - CL*C2 - Q*C2 + Q*C3;
    d/dt(peri)  =                    Q*C2 - Q*C3;
    d/dt(eff)  = Kin - Kout*(1-C2/(EC50+C2))*eff;

Value

None, these functions are used for their side effects.

Note

These functions create a simple, but working example of a dosing regimen simulation web application. Users may want to modify the code to experiment creating shiny applications for their specific rxode2 models.

See Also

rxode2(),eventTable(), and the package shiny (https://shiny.posit.co).

Examples

# remove myapp when the example is complete
on.exit(unlink("myapp", recursive = TRUE, force = TRUE))
# create the shiny app example (template)
genShinyApp.template(appDir = "myapp")
# run the shiny app
if (requireNamespace("shiny", quietly=TRUE)) {
  library(shiny)
  # runApp("myapp") # Won't launch in environments without browsers
}

Get the base simulation model for simulation with inis in the underlying rxode2 model

Description

Get the base simulation model for simulation with inis in the underlying rxode2 model

Usage

getBaseIniSimModel(obj)

Arguments

Value

Simulation object

Author(s)

Matthew L. Fidler

Get the base simulation model for simulation

Description

Get the base simulation model for simulation

Usage

getBaseSimModel(obj)

## Default S3 method:
getBaseSimModel(obj)

## Default S3 method:
getBaseIniSimModel(obj)

Arguments

Value

Simulation object

Author(s)

Matthew L. Fidler

Get the symengine for loading into symengine with rxS()

Description

Get the symengine for loading into symengine with rxS()

Usage

getBaseSymengineModel(obj)

Arguments

Value

Simulation model ready to load into symeng

Author(s)

Matthew L. Fidler

Get/Set the number of threads that rxode2 uses

Description

Get/Set the number of threads that rxode2 uses

Usage

getRxThreads(verbose = FALSE)

setRxThreads(threads = NULL, percent = NULL, throttle = NULL)

rxCores(verbose = FALSE)

Arguments

Value

number of threads that rxode2 uses

Empty Guide

Description

This empty guide draws nothing; It is included in rxode2 for compatibility with ggplot 3.2

Usage

guide_none(title = waiver(), position = waiver())

Arguments

Value

nothing, simply included to be compatible with ggplot 3.2

Assign the ini block in the rxode2 related object

Description

Assign the ini block in the rxode2 related object

Usage

ini(x, envir = environment(x)) <- value

Arguments

Value

rxode2 related object

Author(s)

Matthew L. Fidler

Ini block for rxode2/nlmixr models

Description

The ini block controls initial conditions for 'theta' (fixed effects), 'omega' (random effects), and 'sigma' (residual error) elements of the model.

Usage

## S3 method for class 'rxUi'
ini(x, ..., envir = parent.frame(), append = NULL)

## Default S3 method:
ini(x, ..., envir = parent.frame(), append = NULL)

ini(x, ..., envir = parent.frame(), append = NULL)

Arguments

Details

The ini() function is used in two different ways. The main way that it is used is to set the initial conditions and associated attributes (described below) in a model. The other way that it is used is for updating the initial conditions in a model, often using the pipe operator.

'theta' and 'sigma' can be set using either <- or = such as tvCL <- 1 or equivalently tvCL = 1. 'omega' can be set with a ~ such as etaCL ~ 0.1.

Parameters can be named or unnamed (though named parameters are preferred). A named parameter is set using the name on the left of the assignment while unnamed parameters are set without an assignment operator. tvCL <- 1 would set a named parameter of tvCL to 1. Unnamed parameters are set using just the value, such as 1.

For some estimation methods, lower and upper bounds can be set for 'theta' and 'sigma' values. To set a lower and/or upper bound, use a vector of values. The vector is c(lower, estimate, upper). The vector may be given with just the estimate (estimate), the lower bound and estimate (c(lower, estimate)), or all three (c(lower, estimate, upper)). To set an estimate and upper bound without a lower bound, set the lower bound to -Inf, c(-Inf, estimate, upper). When an estimation method does not support bounds, the bounds will be ignored with a warning.

'omega' values can be set as a single value or as the values of a lower-triangular matrix. The values may be set as either a variance-covariance matrix (the default) or as a correlation matrix for the off-diagonals with the standard deviations on the diagonals. Names may be set on the left side of the ~. To set a variance-covariance matrix with variance values of 2 and 3 and a covariance of -2.5 use ~c(2, 2.5, 3). To set the same matrix with names of iivKa and iivCL, use iivKa + iivCL~c(2, 2.5, 3). To set a correlation matrix with standard deviations on the diagonal, use cor() like iivKa + iivCL~cor(2, -0.5, 3). As of rxode2 3.0 you can also use iivKa ~ 2, iivCL ~ c(2.5, 3) for covariance matrices as well.

Values may be fixed (and therefore not estimated) using either the name fixed at the end of the assignment or by calling fixed() as a function for the value to fix. For 'theta' and 'sigma', either the estimate or the full definition (including lower and upper bounds) may be included in the fixed setting. For example, the following are all effectively equivalent to set a 'theta' or 'sigma' to a fixed value (because the lower and upper bounds are ignored for a fixed value): tvCL <- fixed(1), tvCL <- fixed(0, 1), tvCL <- fixed(0, 1, 2), tvCL <- c(0, fixed(1), 2), or tvCL <- c(0, 1, fixed). For 'omega' assignment, the full block or none of the block must be set as fixed. Examples of setting an 'omega' value as fixed are: iivKa~fixed(1), iivKa + iivCL~fixed(1, 2, 3), or iivKa + iivCL~c(1, 2, 3, fixed). Anywhere that fixed is used, FIX, FIXED, or fix may be used equivalently.

For any value, standard mathematical operators or functions may be used to define the value. For example, log(2) and 24*30 may be used to define a value anywhere that a number can be used (e.g. lower bound, estimate, upper bound, variance, etc.).

Values may be labeled using the label() function after the assignment. Labels are are used to make reporting easier by giving a human-readable description of the parameter, but the labels do not have any effect on estimation. The typical way to set a label so that the parameter tvCL has a label of "Typical Value of Clearance (L/hr)" is tvCL <- 1; label("Typical Value of Clearance (L/hr)").

Off diagonal values of 'omega' can be set to zero using the diag() to remove all off-diagonals can be removed with ini(diag()). To remove covariances of 'omega' item with iivKa, you can use ⁠%>% ini(diag(iivKa))⁠. Or to remove covariances that contain either iivKa or iivCl you can use ⁠%>% ini(diag(iivKa, iivCl))⁠. For finer control you can remove the covariance between two items (like iivKa and iivCl) by '%>% ini(-cov(iivKa, iivCl))

rxode2/nlmixr2 will attempt to determine some back-transformations for the user. For example, CL <- exp(tvCL) will detect that tvCL must be back-transformed by exp() for easier interpretation. When you want to control the back-transformation, you can specify the back-transformation using backTransform() after the assignment. For example, to set the back-transformation to exp(), you can use tvCL <- 1; backTransform(exp()).

Value

ini block

Author(s)

Matthew Fidler

See Also

Other Initial conditions: zeroRe()

Examples

# Set the ini() block in a model
one.compartment <- function() {
  ini({
    tka <- log(1.57); label("Ka")
    tcl <- log(2.72); label("Cl")
    tv <- log(31.5); label("V")
    eta.ka ~ 0.6
    eta.cl ~ 0.3
    eta.v ~ 0.1
    add.sd <- 0.7
  })
  model({
    ka <- exp(tka + eta.ka)
    cl <- exp(tcl + eta.cl)
    v <- exp(tv + eta.v)
    d/dt(depot) = -ka * depot
    d/dt(center) = ka * depot - cl / v * center
    cp = center / v
    cp ~ add(add.sd)
  })
}

# Use piping to update initial conditions
one.compartment %>% ini(tka <- log(2))
one.compartment %>% ini(tka <- label("Absorption rate, Ka (1/hr)"))
# Move the tka parameter to be just below the tv parameter (affects parameter
# summary table, only)
one.compartment %>% ini(tka <- label("Absorption rate, Ka (1/hr)"), append = "tv")
# When programming with rxode2/nlmixr2, it may be easier to pass strings in
# to modify the ini
one.compartment %>% ini("tka <- log(2)")

One correlation sample from the Inverse Wishart distribution

Description

This correlation is constructed by transformation of the Inverse Wishart random covariate to a correlation.

Usage

invWR1d(d, nu, omegaIsChol = FALSE)

Arguments

Value

One correlation sample from the inverse wishart

Author(s)

Matthew Fidler

Check to see if this is an rxEt object.

Description

Check to see if this is an rxEt object.

Usage

is.rxEt(x)

Arguments

Value

Boolean indicating if this is a rxode2 event table

Author(s)

Matthew L.Fidler

Check to see if this is an rxSolve object.

Description

Check to see if this is an rxSolve object.

Usage

is.rxSolve(x)

Arguments

Value

boolean indicating if this is a rxSolve object

Author(s)

Matthew L.Fidler

Return if the object can be stacked

Description

Return if the object can be stacked

Usage

is.rxStackData(object)

Arguments

Value

boolean to tell if an object can be stacked using rxode2

Author(s)

Matthew L. Fidler

Examples

is.rxStackData(NULL)

Leaky ReLU activation function

Description

Leaky ReLU activation function

Usage

lReLU(x)

Arguments

Value

numeric vector

See Also

Other Activation Functions: ELU(), GELU(), PReLU(), ReLU(), SELU(), Swish(), dELU(), dGELU(), dPReLU(), dReLU(), dSELU(), dSwish(), dlReLU(), dsoftplus(), softplus()

Examples

lReLU(c(-1, 0, 1))

# Can use in rxode2 as well

r <- rxode2({r <- lReLU(time)})
e <- et(c(-1, 0, 1))
rxSolve(r, e)

Linear model to replace in rxode2 ui model

Description

Linear model to replace in rxode2 ui model

Usage

linMod(
  variable,
  power,
  dv = "dv",
  intercept = TRUE,
  type = c("replace", "before", "after"),
  num = NULL,
  iniDf = NULL,
  data = FALSE,
  mv = FALSE
)

linMod0(..., intercept = FALSE)

linModB(..., type = "before")

linModB0(..., intercept = FALSE, type = "before")

linModA(..., type = "after")

linModA0(..., intercept = FALSE, type = "after")

linModD(..., intercept = TRUE, data = TRUE)

linModD0(..., intercept = FALSE, data = TRUE)

linModM(..., intercept = TRUE, mv = TRUE)

linModM0(..., intercept = FALSE, mv = TRUE)

Arguments

Value

a list for use in when generating the rxode2 ui model see rxUdfUi() for details.

Functions

• linMod0(): linear model without intercept

• linModB(): linear model before where it occurs

• linModB0(): linear model before where the user function occurs

• linModA(): linear model after where the user function occurs

• linModA0(): liner model without an intercept placed after where the user function occurs

• linModD(): linear model where initial estimates are generated from the data

• linModD0(): linear model where initial estimates are generated from the data (no intercept)

• linModM(): linear model where the model variables are used to generate the model variables

• linModM0(): linear model where the model variables are used to generate the model variables (no intercept)

Author(s)

Matthew L. Fidler

See Also

Other User functions: rxUdfUiControl(), rxUdfUiData(), rxUdfUiEst(), rxUdfUiIniLhs(), rxUdfUiMv(), rxUdfUiNum(), rxUdfUiParsing()

Examples

linMod(x, 3)

Calculate the log likelihood of the binomial function (and its derivatives)

Description

Calculate the log likelihood of the binomial function (and its derivatives)

Usage

llikBeta(x, shape1, shape2, full = FALSE)

Arguments

Details

In an rxode2() model, you can use llikBeta() but you have to use all arguments. You can also get the derivative of shape1 and shape2 with llikBetaDshape1() and llikBetaDshape2().

Value

data frame with fx for the log pdf value of with dShape1 and dShape2 that has the derivatives with respect to the parameters at the observation time-point

Author(s)

Matthew L. Fidler

Examples

x <- seq(1e-4, 1 - 1e-4, length.out = 21)

llikBeta(x, 0.5, 0.5)

llikBeta(x, 1, 3, TRUE)

et <- et(seq(1e-4, 1-1e-4, length.out=21))
et$shape1 <- 0.5
et$shape2 <- 1.5

model <- function() {
  model({
    fx <- llikBeta(time, shape1, shape2)
    dShape1 <- llikBetaDshape1(time, shape1, shape2)
    dShape2 <- llikBetaDshape2(time, shape1, shape2)
  })
}

rxSolve(model, et)

Calculate the log likelihood of the binomial function (and its derivatives)

Description

Calculate the log likelihood of the binomial function (and its derivatives)

Usage

llikBinom(x, size, prob, full = FALSE)

Arguments

Details

In an rxode2() model, you can use llikBinom() but you have to use all arguments. You can also get the derivative of prob with llikBinomDprob()

Value

data frame with fx for the pdf value of with dProb that has the derivatives with respect to the parameters at the observation time-point

Author(s)

Matthew L. Fidler

Examples

llikBinom(46:54, 100, 0.5)

llikBinom(46:54, 100, 0.5, TRUE)

# In rxode2 you can use:

et <- et(46:54)
et$size <- 100
et$prob <-0.5

model <- function() {
  model({
    fx <- llikBinom(time, size, prob)
    dProb <- llikBinomDprob(time, size, prob)
  })
}

rxSolve(model, et)

log likelihood of Cauchy distribution and it's derivatives (from stan)

Description

log likelihood of Cauchy distribution and it's derivatives (from stan)

Usage

llikCauchy(x, location = 0, scale = 1, full = FALSE)

Arguments

Details

In an rxode2() model, you can use llikCauchy() but you have to use all arguments. You can also get the derivative of location and scale with llikCauchyDlocation() and llikCauchyDscale().

Value

data frame with fx for the log pdf value of with dLocation and dScale that has the derivatives with respect to the parameters at the observation time-point

Author(s)

Matthew L. Fidler

Examples

x <- seq(-3, 3, length.out = 21)

llikCauchy(x, 0, 1)

llikCauchy(x, 3, 1, full=TRUE)

et <- et(-3, 3, length.out=10)
et$location <- 0
et$scale <- 1

model <- function() {
  model({
    fx <- llikCauchy(time, location, scale)
    dLocation <- llikCauchyDlocation(time, location, scale)
    dScale <- llikCauchyDscale(time, location, scale)
  })
}

rxSolve(model, et)

log likelihood and derivatives for chi-squared distribution

Description

log likelihood and derivatives for chi-squared distribution

Usage

llikChisq(x, df, full = FALSE)

Arguments

Details

In an rxode2() model, you can use llikChisq() but you have to use the x and df arguments. You can also get the derivative of df with llikChisqDdf().

Value

data frame with fx for the log pdf value of with dDf that has the derivatives with respect to the df parameter the observation time-point

Author(s)

Matthew L. Fidler

Examples

llikChisq(1, df = 1:3, full=TRUE)

llikChisq(1, df = 6:9)

et <- et(1:3)
et$x <- 1

model <- function() {
  model({
   fx <- llikChisq(x, time)
   dDf <- llikChisqDdf(x, time)
  })
}

rxSolve(model, et)

log likelihood and derivatives for exponential distribution

Description

log likelihood and derivatives for exponential distribution

Usage

llikExp(x, rate, full = FALSE)

Arguments

Details

In an rxode2() model, you can use llikExp() but you have to use the x and rate arguments. You can also get the derivative of rate with llikExpDrate().

Value

data frame with fx for the log pdf value of with dRate that has the derivatives with respect to the rate parameter the observation time-point

Author(s)

Matthew L. Fidler

Examples

llikExp(1, 1:3)

llikExp(1, 1:3, full=TRUE)

# You can use rxode2 for these too:

et <- et(1:3)
et$x <- 1

model <- function() {
  model({
    fx <- llikExp(x, time)
    dRate <- llikExpDrate(x, time)
  })
}

rxSolve(model, et)

log likelihood and derivatives for F distribution

Description

log likelihood and derivatives for F distribution

Usage

llikF(x, df1, df2, full = FALSE)

Arguments

Details

In an rxode2() model, you can use llikF() but you have to use the x and rate arguments. You can also get the derivative of df1 and df2 with llikFDdf1() and llikFDdf2().

Value

data frame with fx for the log pdf value of with dDf1 and dDf2 that has the derivatives with respect to the df1/df2 parameters at the observation time-point

Author(s)

Matthew L. Fidler

Examples

x <- seq(0.001, 5, length.out = 100)

llikF(x^2, 1, 5)

model <- function(){
  model({
    fx <- llikF(time, df1, df2)
    dMean <- llikFDdf1(time, df1, df2)
    dSd <- llikFDdf2(time, df1, df2)
  })
}

et <- et(x)
et$df1 <- 1
et$df2 <- 5

rxSolve(model, et)

log likelihood and derivatives for Gamma distribution

Description

log likelihood and derivatives for Gamma distribution

Usage

llikGamma(x, shape, rate, full = FALSE)

Arguments

Details

In an rxode2() model, you can use llikGamma() but you have to use the x and rate arguments. You can also get the derivative of shape or rate with llikGammaDshape() and llikGammaDrate().

Value

data frame with fx for the log pdf value of with dProb that has the derivatives with respect to the prob parameters at the observation time-point

Author(s)

Matthew L. Fidler

Examples

llikGamma(1, 1, 10)

# You can use this in `rxode2` too:

et  <- et(seq(0.001, 1, length.out=10))
et$shape <- 1
et$rate <- 10
 
model <- function() {
  model({
    fx <- llikGamma(time, shape, rate)
    dShape<- llikGammaDshape(time, shape, rate)
    dRate <- llikGammaDrate(time, shape, rate)
  })
}

rxSolve(model, et)

log likelihood and derivatives for Geom distribution

Description

log likelihood and derivatives for Geom distribution

Usage

llikGeom(x, prob, full = FALSE)

Arguments

Details

In an rxode2() model, you can use llikGeom() but you have to use the x and rate arguments. You can also get the derivative of prob with llikGeomDprob().

Value

data frame with fx for the log pdf value of with dProb that has the derivatives with respect to the prob parameters at the observation time-point

Author(s)

Matthew L. Fidler

Examples

llikGeom(1:10, 0.2)

et  <- et(1:10)
et$prob <- 0.2
 
model <- function() {
  model({
    fx <- llikGeom(time, prob)
    dProb <- llikGeomDprob(time, prob)
  })
}

rxSolve(model, et)

Calculate the log likelihood of the negative binomial function (and its derivatives)

Description

Calculate the log likelihood of the negative binomial function (and its derivatives)

Usage

llikNbinom(x, size, prob, full = FALSE)

Arguments

Details

In an rxode2() model, you can use llikNbinom() but you have to use all arguments. You can also get the derivative of prob with llikNbinomDprob()

Value

data frame with fx for the pdf value of with dProb that has the derivatives with respect to the parameters at the observation time-point

Author(s)

Matthew L. Fidler

Examples

llikNbinom(46:54, 100, 0.5)

llikNbinom(46:54, 100, 0.5, TRUE)

# In rxode2 you can use:

et <- et(46:54)
et$size <- 100
et$prob <-0.5

model <- function() {
  model({
    fx <- llikNbinom(time, size, prob)
    dProb <- llikNbinomDprob(time, size, prob)
  })
}

rxSolve(model, et)

Calculate the log likelihood of the negative binomial function (and its derivatives)

Description

Calculate the log likelihood of the negative binomial function (and its derivatives)

Usage

llikNbinomMu(x, size, mu, full = FALSE)

Arguments

Details

In an rxode2() model, you can use llikNbinomMu() but you have to use all arguments. You can also get the derivative of mu with llikNbinomMuDmu()

Value

data frame with fx for the pdf value of with dProb that has the derivatives with respect to the parameters at the observation time-point

Author(s)

Matthew L. Fidler

Examples

llikNbinomMu(46:54, 100, 40)

llikNbinomMu(46:54, 100, 40, TRUE)

et <- et(46:54)
et$size <- 100
et$mu <- 40

model <- function() {
  model({
    fx <- llikNbinomMu(time, size, mu)
    dProb <- llikNbinomMuDmu(time, size, mu)
  })
}

rxSolve(model, et)

Log likelihood for normal distribution

Description

Log likelihood for normal distribution

Usage

llikNorm(x, mean = 0, sd = 1, full = FALSE)

Arguments

Details

In an rxode2() model, you can use llikNorm() but you have to use all arguments. You can also get the derivatives with llikNormDmean() and llikNormDsd()

Value

data frame with fx for the pdf value of with dMean and dSd that has the derivatives with respect to the parameters at the observation time-point

Author(s)

Matthew L. Fidler

Examples

llikNorm(0)

llikNorm(seq(-2,2,length.out=10), full=TRUE)

# With rxode2 you can use:
 
et <- et(-3, 3, length.out=10)
et$mu <- 0
et$sigma <- 1

model <- function(){
  model({
    fx <- llikNorm(time, mu, sigma)
    dMean <- llikNormDmean(time, mu, sigma)
    dSd <- llikNormDsd(time, mu, sigma)
  })
 }

ret <- rxSolve(model, et)
ret

log-likelihood for the Poisson distribution

Description

log-likelihood for the Poisson distribution

Usage

llikPois(x, lambda, full = FALSE)

Arguments

Details

In an rxode2() model, you can use llikPois() but you have to use all arguments. You can also get the derivatives with llikPoisDlambda()

Value

data frame with fx for the pdf value of with dLambda that has the derivatives with respect to the parameters at the observation time-point

Author(s)

Matthew L. Fidler

Examples

llikPois(0:7, lambda = 1)

llikPois(0:7, lambda = 4, full=TRUE)

# In rxode2 you can use:

et <- et(0:10)
et$lambda <- 0.5

model <- function() {
  model({
    fx <- llikPois(time, lambda)
    dLambda <- llikPoisDlambda(time, lambda)
  })
}

rxSolve(model, et)

Log likelihood of T and it's derivatives (from stan)

Description

Log likelihood of T and it's derivatives (from stan)

Usage

llikT(x, df, mean = 0, sd = 1, full = FALSE)

Arguments

Details

In an rxode2() model, you can use llikT() but you have to use all arguments. You can also get the derivative of df, mean and sd with llikTDdf(), llikTDmean() and llikTDsd().

Value

data frame with fx for the log pdf value of with dDf dMean and dSd that has the derivatives with respect to the parameters at the observation time-point

Author(s)

Matthew L. Fidler

Examples

x <- seq(-3, 3, length.out = 21)

llikT(x, 7, 0, 1)

llikT(x, 15, 0, 1, full=TRUE)

et <- et(-3, 3, length.out=10)
et$nu <- 7
et$mean <- 0
et$sd <- 1

model <- function() {
  model({
    fx <- llikT(time, nu, mean, sd)
    dDf <- llikTDdf(time, nu, mean, sd)
    dMean <- llikTDmean(time, nu, mean, sd)
    dSd   <- llikTDsd(time, nu, mean, sd)
  })
}

rxSolve(model, et)

log likelihood and derivatives for Unif distribution

Description

log likelihood and derivatives for Unif distribution

Usage

llikUnif(x, alpha, beta, full = FALSE)

Arguments

Details

In an rxode2() model, you can use llikUnif() but you have to use the x and rate arguments. You can also get the derivative of alpha or beta with llikUnifDalpha() and llikUnifDbeta().

Value

data frame with fx for the log pdf value of with dProb that has the derivatives with respect to the prob parameters at the observation time-point

Author(s)

Matthew L. Fidler

Examples

llikUnif(1, -2, 2)

et  <- et(seq(1,1, length.out=4))
et$alpha <- -2
et$beta <- 2
 
model <- function() {
  model({
    fx <- llikUnif(time, alpha, beta)
    dAlpha<- llikUnifDalpha(time, alpha, beta)
    dBeta <- llikUnifDbeta(time, alpha, beta)
  })
}

rxSolve(model, et)

log likelihood and derivatives for Weibull distribution

Description

log likelihood and derivatives for Weibull distribution

Usage

llikWeibull(x, shape, scale, full = FALSE)

Arguments

Details

In an rxode2() model, you can use llikWeibull() but you have to use the x and rate arguments. You can also get the derivative of shape or scale with llikWeibullDshape() and llikWeibullDscale().

Value

data frame with fx for the log pdf value of with dProb that has the derivatives with respect to the prob parameters at the observation time-point

Author(s)

Matthew L. Fidler

Examples

llikWeibull(1, 1, 10)

# rxode2 can use this too:

et  <- et(seq(0.001, 1, length.out=10))
et$shape <- 1
et$scale <- 10
 
model <- function() {
  model({
    fx <- llikWeibull(time, shape, scale)
    dShape<- llikWeibullDshape(time, shape, scale)
    dScale <- llikWeibullDscale(time, shape, scale)
  })
}

rxSolve(model, et)

logit and inverse logit (expit) functions

Description

logit and inverse logit (expit) functions

Usage

logit(x, low = 0, high = 1)

expit(alpha, low = 0, high = 1)

logitNormInfo(mean = 0, sd = 1, low = 0, high = 1, abs.tol = 1e-06, ...)

probitNormInfo(mean = 0, sd = 1, low = 0, high = 1, abs.tol = 1e-06, ...)

Arguments

Details

logit is given by:

logit(p) = -log(1/p-1)

where:

p = x-low/high-low

expit is given by:

expit(p, low, high) = (high-low)/(1+exp(-alpha)) + low

The logitNormInfo() gives the mean, variance and coefficient of variability on the untransformed scale.

Value

values from logit and expit

Examples

logit(0.25)

expit(-1.09)

logitNormInfo(logit(0.25), sd = 0.1)

logitNormInfo(logit(1, 0, 10), sd = 1, low = 0, high = 10)

lowergamma: upper incomplete gamma function

Description

This is the tgamma_lower from the boost library

Usage

lowergamma(a, z)

Arguments

Details

The lowergamma function is given by:

lowergamma(a, z) = \int_{0}^{z}t^{a-1}\cdot e^{-t} dt

Value

lowergamma results

Author(s)

Matthew L. Fidler

Examples

lowergamma(1, 3)

lowergamma(1:3, 3)

lowergamma(1, 1:3)

Calculate expected confidence bands or prediction intreval with normal or t sampling distribution

Description

The generic function meanProbs produces expected confidence bands under either the t distribution or the normal sampling distribution. This uses qnorm() or qt() with the mean and standard deviation.

Usage

meanProbs(x, ...)

## Default S3 method:
meanProbs(
  x,
  probs = seq(0, 1, 0.25),
  na.rm = FALSE,
  names = TRUE,
  useT = TRUE,
  onlyProbs = TRUE,
  pred = FALSE,
  n = 0L,
  ...
)

Arguments

Details

For a single probability, p, it uses either:

mean + qt(p, df=n)*sd/sqrt(n)

or

mean + qnorm(p)*sd/sqrt(n)

The smallest observation corresponds to a probability of 0 and the largest to a probability of 1 and the mean corresponds to 0.5.

The mean and standard deviation of the sample is calculated based on Welford's method for a single pass.

This is meant to perform in the same way as quantile() so it can be a drop in replacement for code using quantile() but using distributional assumptions.

Value

By default the return has the probabilities as names (if named) with the points where the expected distribution are located given the sampling mean and standard deviation. If onlyProbs=FALSE then it would prepend mean, variance, standard deviation, minimum, maximum and number of non-NA observations.

Author(s)

Matthew L. Fidler

Examples

quantile(x<- rnorm(1001))
meanProbs(x)

# Can get some extra statistics if you request onlyProbs=FALSE
meanProbs(x, onlyProbs=FALSE)

x[2] <- NA_real_

meanProbs(x, onlyProbs=FALSE)

quantile(x<- rnorm(42))

meanProbs(x)

meanProbs(x, useT=FALSE)

Assign the model block in the rxode2 related object

Description

Assign the model block in the rxode2 related object

Usage

model(x, envir = environment(x)) <- value

Arguments

Value

rxode2 related object

Author(s)

Matthew L. Fidler

Model block for rxode2/nlmixr models

Description

Model block for rxode2/nlmixr models

Usage

## S3 method for class ''function''
model(
  x,
  ...,
  append = NULL,
  auto = getOption("rxode2.autoVarPiping", TRUE),
  cov = NULL,
  envir = parent.frame()
)

## S3 method for class 'rxUi'
model(
  x,
  ...,
  append = NULL,
  auto = getOption("rxode2.autoVarPiping", TRUE),
  cov = NULL,
  envir = parent.frame()
)

## S3 method for class 'rxode2'
model(
  x,
  ...,
  append = NULL,
  auto = getOption("rxode2.autoVarPiping", TRUE),
  cov = NULL,
  envir = parent.frame()
)

## S3 method for class 'rxModelVars'
model(
  x,
  ...,
  append = NULL,
  auto = getOption("rxode2.autoVarPiping", TRUE),
  cov = NULL,
  envir = parent.frame()
)

model(
  x,
  ...,
  append = FALSE,
  auto = getOption("rxode2.autoVarPiping", TRUE),
  cov = NULL,
  envir = parent.frame()
)

## Default S3 method:
model(x, ..., append = FALSE, cov = NULL, envir = parent.frame())

Arguments

Value

Model block with ini information included. ini must be called before model block

Author(s)

Matthew Fidler

Extract model lines from a rxui model

Description

Extract model lines from a rxui model

Usage

modelExtract(
  x,
  ...,
  expression = FALSE,
  endpoint = FALSE,
  lines = FALSE,
  envir = parent.frame()
)

## S3 method for class ''function''
modelExtract(
  x,
  ...,
  expression = FALSE,
  endpoint = FALSE,
  lines = FALSE,
  envir = parent.frame()
)

## S3 method for class 'rxUi'
modelExtract(
  x,
  ...,
  expression = FALSE,
  endpoint = FALSE,
  lines = FALSE,
  envir = parent.frame()
)

## S3 method for class 'rxode2'
modelExtract(
  x,
  ...,
  expression = FALSE,
  endpoint = FALSE,
  lines = FALSE,
  envir = parent.frame()
)

## S3 method for class 'rxModelVars'
modelExtract(
  x,
  ...,
  expression = FALSE,
  endpoint = FALSE,
  lines = FALSE,
  envir = parent.frame()
)

## Default S3 method:
modelExtract(
  x,
  ...,
  expression = FALSE,
  endpoint = FALSE,
  lines = FALSE,
  envir = parent.frame()
)

Arguments

Value

expressions or strings of extracted lines. Note if there is a duplicated lhs expression in the line, it will return both lines

Author(s)

Matthew L. Fidler

Examples

one.compartment <- function() {
  ini({
    tka <- 0.45 # Log Ka
    tcl <- 1 # Log Cl
    tv <- 3.45    # Log V
    eta.ka ~ 0.6
    eta.cl ~ 0.3
    eta.v ~ 0.1
    add.sd <- 0.7
  })
  model({
    ka <- exp(tka + eta.ka)
    cl <- exp(tcl + eta.cl)
    v  <- exp(tv + eta.v)
    d/dt(depot)  <- -ka * depot
    d/dt(center) <-  ka * depot - cl / v * center
    cp <- center / v
    cp ~ add(add.sd)
  })
 }

 f <- one.compartment()

 modelExtract(f, cp)

 modelExtract(one.compartment, d/dt(depot))

 # from variable
 var <- "d/dt(depot)"

 modelExtract(one.compartment, var)

 modelExtract(f, endpoint=NA, lines=TRUE, expression=TRUE)

Conversion between character and integer ODE integration methods for rxode2

Description

If NULL is given as the method, all choices are returned as a named vector.

Usage

odeMethodToInt(method = c("liblsoda", "lsoda", "dop853", "indLin"))

Arguments

Value

An integer for the method (unless the input is NULL, in which case, see the details)

Cumulative distribution of standard normal

Description

Cumulative distribution of standard normal

Usage

phi(q)

Arguments

Value

cumulative distribution of standard normal distribution

Author(s)

Matthew Fidler

Examples

# phi is equivalent to pnorm(x)
phi(3)

# See
pnorm(3)

# This is provided for NONMEM-like compatibility in rxode2 models

Plot rxode2 objects

Description

Plot rxode2 objects

Usage

## S3 method for class 'rxSolve'
plot(x, y, ..., log = "", xlab = "Time", ylab = "")

## S3 method for class 'rxSolveConfint1'
plot(x, y, ..., xlab = "Time", ylab = "", log = "")

## S3 method for class 'rxSolveConfint2'
plot(x, y, ..., xlab = "Time", ylab = "", log = "")

Arguments

Value

A ggplot2 object

See Also

Other rxode2 plotting: rxTheme()

Print the rxCoef object

Description

This prints out the user supplied arguments for rxCoef object

Usage

## S3 method for class 'rxCoef'
print(x, ...)

Arguments

Value

original object

Author(s)

Matthew L.Fidler

Print rxDll object

Description

This tells if the rxDll is loaded, ready and/or deleted.

Usage

## S3 method for class 'rxDll'
print(x, ...)

Value

original object

Author(s)

Matthew L.Fidler

Print Values

Description

print prints its argument and returns it invisibly (via invisible(x)). It is a generic function which means that new printing methods can be easily added for new classes.

Usage

## S3 method for class 'rxModelVars'
print(x, ...)

Arguments

Details

The default method, print.default has its own help page. Use methods("print") to get all the methods for the print generic.

print.factor allows some customization and is used for printing ordered factors as well.

print.table for printing tables allows other customization. As of R 3.0.0, it only prints a description in case of a table with 0-extents (this can happen if a classifier has no valid data).

See noquote as an example of a class whose main purpose is a specific print method.

Value

This returns invisibly the model variables object

References

Chambers, J. M. and Hastie, T. J. (1992) Statistical Models in S. Wadsworth & Brooks/Cole.

See Also

The default method print.default, and help for the methods above; further options, noquote.

For more customizable (but cumbersome) printing, see cat, format or also write. For a simple prototypical print method, see .print.via.format in package tools.

Examples

require(stats)

ts(1:20)  #-- print is the "Default function" --> print.ts(.) is called
for(i in 1:3) print(1:i)

## Printing of factors
attenu$station ## 117 levels -> 'max.levels' depending on width

## ordered factors: levels  "l1 < l2 < .."
esoph$agegp[1:12]
esoph$alcgp[1:12]

## Printing of sparse (contingency) tables
set.seed(521)
t1 <- round(abs(rt(200, df = 1.8)))
t2 <- round(abs(rt(200, df = 1.4)))
table(t1, t2) # simple
print(table(t1, t2), zero.print = ".") # nicer to read

## same for non-integer "table":
T <- table(t2,t1)
T <- T * (1+round(rlnorm(length(T)))/4)
print(T, zero.print = ".") # quite nicer,
print.table(T[,2:8] * 1e9, digits=3, zero.print = ".")
## still slightly inferior to  Matrix::Matrix(T)  for larger T

## Corner cases with empty extents:
table(1, NA) # < table of extent 1 x 0 >

Print information about the rxode2 object.

Description

This prints the model name and its status for being able to be solved

Usage

## S3 method for class 'rxode2'
print(x, ...)

Arguments

Value

original object

Author(s)

Matthew L.Fidler

probit and inverse probit functions

Description

probit and inverse probit functions

Usage

probit(x, low = 0, high = 1)

probitInv(x, low = 0, high = 1)

Arguments

Value

values from probit, probitInv and probitNormInfo

Examples

probit(0.25)

probitInv(-0.674)

probitNormInfo(probit(0.25), sd = 0.1)

probitNormInfo(probit(1, 0, 10), sd = 1, low = 0, high = 10)

One correlation sample from the LKJ distribution

Description

One correlation sample from the LKJ distribution

Usage

rLKJ1(d, eta = 1, cholesky = FALSE)

Arguments

Value

A correlation sample from the LKJ distribution

Author(s)

Matthew Fidler (translated to RcppArmadillo) and Emma Schwager

Objects exported from other packages

Description

These objects are imported from other packages. Follow the links below to see their documentation.

ggplot2

aes, expand_limits, facet_wrap, geom_line, ggplot, label_both, label_context, label_context, label_value, label_wrap_gen, scale_type, scale_x_continuous, scale_x_date, scale_x_discrete, scale_y_continuous, scale_y_date, scale_y_discrete, waiver, xlab, ylab

lotri

lotri

magrittr

%>%

Value

Inherited from parent routine

Scaled Inverse Chi Squared distribution

Description

Scaled Inverse Chi Squared distribution

Usage

rinvchisq(n = 1L, nu = 1, scale = 1)

Arguments

Value

a vector of inverse chi squared deviates.

Examples

rinvchisq(3, 4, 1) ## Scale = 1, degrees of freedom = 4
rinvchisq(2, 4, 2) ## Scale = 2, degrees of freedom = 4

Allow unloading of dlls

Description

Allow unloading of dlls

Usage

rxAllowUnload(allow)

Arguments

Value

Boolean allow; called for side effects

Author(s)

Matthew Fidler

Examples

# Garbage collection will not unload un-used rxode2 dlls
rxAllowUnload(FALSE);

# Garbage collection will unload unused rxode2 dlls
rxAllowUnload(TRUE);

Append two rxui models together

Description

Append two rxui models together

Usage

rxAppendModel(..., common = TRUE)

Arguments

Value

New model with both models appended together

Author(s)

Matthew L. Fidler

Examples

ocmt <- function() {
  ini({
    tka <- exp(0.45) # Ka
    tcl <- exp(1) # Cl
    tv <- exp(3.45); # log V
    ## the label("Label name") works with all models
    add.sd <- 0.7
  })
  model({
    ka <- tka
    cl <- tcl
    v <- tv
    d/dt(depot) <- -ka * depot
    d/dt(center) <- ka * depot - cl / v * center
    cp <- center / v
    cp ~ add(add.sd)
  })
}

idr <- function() {
  ini({
    tkin <- log(1)
    tkout <- log(1)
    tic50 <- log(10)
    gamma <- fix(1)
    idr.sd <- 1
  })
  model({
    kin <- exp(tkin)
    kout <- exp(tkout)
    ic50 <- exp(tic50)
    d/dt(eff) <- kin - kout*(1-ceff^gamma/(ic50^gamma+ceff^gamma))
    eff ~ add(idr.sd)
  })
}

rxAppendModel(ocmt %>% model(ceff=cp,append=TRUE), idr)

Assign Control Variable

Description

Assign Control Variable

Usage

rxAssignControlValue(ui, option, value)

Arguments

Value

Nothing; called for the side effects

Author(s)

Matthew L. Fidler

Assign pointer based on model variables

Description

Assign pointer based on model variables

Usage

rxAssignPtr(object = NULL)

Arguments

Value

nothing, called for side effects

Creates a logical matrix for block matrixes.

Description

Creates a logical matrix for block matrixes.

Usage

rxBlockZeros(mat, i)

Arguments

Value

A logical matrix returning where the elements should be zero.

Return the C file associated with the rxode2 object

Description

This will return C code for generating the rxode2 DLL.

Usage

rxC(obj)

Arguments

Value

a path of the library

Author(s)

Matthew L.Fidler

Use cat when rxode2.verbose is TRUE

Description

Use cat when rxode2.verbose is TRUE

Usage

rxCat(a, ...)

Arguments

Value

nothing

Author(s)

Matthew L. Fidler

Bind the study parameters and individual parameters

Description

Bind the study parameters and individual parameters

Usage

rxCbindStudyIndividual(studyParameters, individualParameters)

Arguments

Value

Data frame that can be used in rxode2 simulations

Author(s)

Matthew Fidler

Examples

# Function for coverting coefficient of covariance into a variance
lognCv <- function(x){log((x/100)^2+1)}

set.seed(32)

nSub  <-  100
nStud  <-  10

#define theta
theta <- c(lka=log(0.5), # log ka
          lCl=log(5), # log Cl
          lV=log(300) # log V
          )

#define theta Matrix
thetaMat <- lotri(lCl ~ lognCv(5),
                 lV  ~ lognCv(5),
                 lka ~ lognCv(5))


nev <- nSub*nStud

ev1 <- data.frame(COV1=rnorm(nev,50,30),COV2=rnorm(nev,75,10),
                  COV3=sample(c(1.0,2.0),nev,replace=TRUE))


tmat <-rxRmvn(nStud, theta[dimnames(thetaMat)[[1]]], thetaMat)

rxCbindStudyIndividual(tmat, ev1)

rxChain Chain or add item to solved system of equations

Description

Add item to solved system of equations

Usage

rxChain(obj1, obj2)

## S3 method for class 'solveRxDll'
obj1 + obj2

Arguments

Value

When newObject is an event table, return a new solved object with the new event table.

Author(s)

Matthew L. Fidler

Second command in chaining commands

Description

This is s3 method is called internally with + and ⁠\%>\%⁠ operators.

Usage

rxChain2(obj, solvedObject)

## Default S3 method:
rxChain2(obj, solvedObject)

## S3 method for class 'EventTable'
rxChain2(obj, solvedObject)

Arguments

Value

chained operation

Author(s)

Matthew L.Fidler

Cleanup anonymous DLLs by unloading them

Description

This cleans up any rxode2 loaded DLLs

Usage

rxClean(wd)

Arguments

Value

TRUE if successful

Author(s)

Matthew L. Fidler

Combine Error Lines and create rxode2 expression

Description

Combine Error Lines and create rxode2 expression

Usage

rxCombineErrorLines(
  uiModel,
  errLines = NULL,
  prefixLines = NULL,
  paramsLine = NULL,
  modelVars = FALSE,
  cmtLines = TRUE,
  dvidLine = TRUE,
  lstExpr = NULL,
  useIf = TRUE,
  interpLines = NULL,
  levelLines = NULL
)

Arguments

Details

This is exported to allow other functions to mangle the error lines to make other types of estimation methods (if needed)

Value

quoted expression that can be evaluated to compiled rxode2 model

Author(s)

Matthew L. Fidler

Examples

one.cmt <- function() {
  ini({
    ## You may label each parameter with a comment
    tka <- 0.45 # Log Ka
    tcl <- log(c(0, 2.7, 100)) # Log Cl
    ## This works with interactive models
    ## You may also label the preceding line with label("label text")
    tv <- 3.45; label("log V")
    ## the label("Label name") works with all models
    eta.ka ~ 0.6
    eta.cl ~ 0.3
    eta.v ~ 0.1
    add.sd <- 0.7
  })
  model({
    ka <- exp(tka + eta.ka)
    cl <- exp(tcl + eta.cl)
    v <- exp(tv + eta.v)
    linCmt() ~ add(add.sd)
  })
}

f <- rxode2(one.cmt)

# You can get the simulation model easily by
rxCombineErrorLines(f)

# You can then get the compiled model by simply evaluting the model:
r <- eval(rxCombineErrorLines(f))

# This also works with multile endpoint models:
pk.turnover.emax <- function() {
  ini({
    tktr <- log(1)
    tka <- log(1)
    tcl <- log(0.1)
    tv <- log(10)
    ##
    eta.ktr ~ 1
    eta.ka ~ 1
    eta.cl ~ 2
    eta.v ~ 1
    prop.err <- 0.1
    pkadd.err <- 0.1
    ##
    temax <- logit(0.8)
    tec50 <- log(0.5)
    tkout <- log(0.05)
    te0 <- log(100)
    ##
    eta.emax ~ .5
    eta.ec50  ~ .5
    eta.kout ~ .5
    eta.e0 ~ .5
    ##
    pdadd.err <- 10
  })
  model({
    ktr <- exp(tktr + eta.ktr)
    ka <- exp(tka + eta.ka)
    cl <- exp(tcl + eta.cl)
    v <- exp(tv + eta.v)
    ##
    emax=expit(temax+eta.emax)
    ec50 =  exp(tec50 + eta.ec50)
    kout = exp(tkout + eta.kout)
    e0 = exp(te0 + eta.e0)
    ##
    DCP = center/v
    PD=1-emax*DCP/(ec50+DCP)
    ##
    effect(0) = e0
    kin = e0*kout
    ##
    d/dt(depot) = -ktr * depot
    d/dt(gut) =  ktr * depot -ka * gut
    d/dt(center) =  ka * gut - cl / v * center
    d/dt(effect) = kin*PD -kout*effect
    ##
    cp = center / v
    cp ~ prop(prop.err) + add(pkadd.err)
    effect ~ add(pdadd.err)
  })
}

f <- rxode2(pk.turnover.emax)
rxCombineErrorLines(f)

# Note that in the parsed form, you can also get the compiled rxode2
# model with $simulationModel

f$simulationModel

Compile a model if needed

Description

This is the compilation workhorse creating the rxode2 model DLL files.

Usage

rxCompile(
  model,
  dir,
  prefix,
  force = FALSE,
  modName = NULL,
  package = NULL,
  ...
)

## S3 method for class 'rxModelVars'
rxCompile(
  model,
  dir = NULL,
  prefix = NULL,
  force = FALSE,
  modName = NULL,
  package = NULL,
  ...
)

## S3 method for class 'character'
rxCompile(
  model,
  dir = NULL,
  prefix = NULL,
  force = FALSE,
  modName = NULL,
  package = NULL,
  ...
)

## S3 method for class 'rxDll'
rxCompile(model, ...)

## S3 method for class 'rxode2'
rxCompile(model, ...)

Arguments

If unspecified, the C code is stored in a temporary directory,
then the model is compiled and moved to the current directory.
Afterwards the C code is removed.

If specified, the C code is stored in the specified directory
and then compiled in that directory.  The C code is not removed
after the DLL is created in the same directory.  This can be
useful to debug the c-code outputs.

Value

An rxDll object that has the following components

• dll DLL path

• model model specification

• .c A function to call C code in the correct context from the DLL using the .C() function.

• .call A function to call C code in the correct context from the DLL using the .Call() function.

• args A list of the arguments used to create the rxDll object.

Author(s)

Matthew L.Fidler

See Also

rxode2()

Current Condition for rxode2 object

Description

Current Condition for rxode2 object

Usage

rxCondition(obj, condition = NULL)

Arguments

Value

Current condition for rxode2 object

Author(s)

Matthew L. Fidler

This updates the tolerances based on the sensitivity equations

Description

This assumes the normal ODE equations are the first equations and the ODE is expanded by the forward sensitivities or other type of sensitivity (like adjoint)

Usage

rxControlUpdateSens(rxControl, sensCmt = NULL, ncmt = NULL)

Arguments

Value

Updated rxControl where ⁠$atol⁠, ⁠$rtol⁠, ⁠$ssAtol⁠ ⁠$ssRtol⁠ are updated with different sensitivities for the normal ODEs (first) and a different sensitivity for the larger compartments (sensitivities).

Author(s)

Matthew L. Fidler

Examples

tmp <- rxControl()

tmp2 <- rxControlUpdateSens(tmp, 3, 6)

tmp2$atol
tmp2$rtol
tmp2$ssAtol
tmp2$ssRtol

This will create the cache directory for rxode2 to save between sessions

Description

When run, if the R_user_dir for rxode2's cache isn't present, create the cache

Usage

rxCreateCache()

Value

nothing

Author(s)

Matthew Fidler

Add to rxode2's derivative tables

Description

Add to rxode2's derivative tables

Usage

rxD(name, derivatives)

Arguments

Value

nothing

Author(s)

Matthew Fidler

Examples

## Add an arbitrary list of derivative functions
## In this case the fun(x,y) is assumed to be 0.5*x^2+0.5*y^2

rxD("fun", list(
  function(x, y) {
    return(x)
  },
  function(x, y) {
    return(y)
  }
))

Delete the DLL for the model

Description

This function deletes the DLL, but doesn't delete the model information in the object.

Usage

rxDelete(obj)

Arguments

Value

A boolean stating if the operation was successful.

Author(s)

Matthew L.Fidler

Demote the error type

Description

Demote the error type

Usage

rxDemoteAddErr(errType)

Arguments

Value

Demoted Error Type

Author(s)

Matthew Fidler

Examples

rxErrTypeCombine("add") %>%
  rxErrTypeCombine("prop")

# This removes the internal additive error
rxErrTypeCombine("add") %>%
  rxErrTypeCombine("prop") %>%
  rxDemoteAddErr()

# This is used for logitNorm(NA), the additive portion is stripped

Calculate derived parameters for the 1-, 2-, and 3- compartment linear models.

Description

This calculates the derived parameters based on what is provided in a data frame or arguments

Usage

rxDerived(..., verbose = FALSE, digits = 0)

Arguments

Value

Return a data.frame of derived PK parameters for a 1-, 2-, or 3-compartment linear model given provided clearances and volumes based on the inferred model type.

The model parameters that will be provided in the data frame are:

• vc: Central Volume (for 1-, 2- and 3- compartment models)

• kel: First-order elimination rate (for 1-, 2-, and 3-compartment models)

• k12: First-order rate of transfer from central to first peripheral compartment; (for 2- and 3-compartment models)

• k21: First-order rate of transfer from first peripheral to central compartment, (for 2- and 3-compartment models)

• k13: First-order rate of transfer from central to second peripheral compartment; (3-compartment model)

• k31: First-order rate of transfer from second peripheral to central compartment (3-compartment model)

• vp: Peripheral Volume (for 2- and 3- compartment models)

• vp2: Peripheral Volume for 3rd compartment (3- compartment model)

• vss: Volume of distribution at steady state; (1-, 2-, and 3-compartment models)

• t12alpha: t_{1/2,\alpha}; (1-, 2-, and 3-compartment models)

• t12beta: t_{1/2,\beta}; (2- and 3-compartment models)

• t12gamma: t_{1/2,\gamma}; (3-compartment model)

• alpha: \alpha; (1-, 2-, and 3-compartment models)

• beta: \beta; (2- and 3-compartment models)

• gamma: \beta; (3-compartment model)

• A: true A; (1-, 2-, and 3-compartment models)

• B: true B; (2- and 3-compartment models)

• C: true C; (3-compartment model)

• fracA: fractional A; (1-, 2-, and 3-compartment models)

• fracB: fractional B; (2- and 3-compartment models)

• fracC: fractional C; (3-compartment model)

Author(s)

Matthew Fidler and documentation from Justin Wilkins, justin.wilkins@occams.com

References

Shafer S. L. CONVERT.XLS

Rowland M, Tozer TN. Clinical Pharmacokinetics and Pharmacodynamics: Concepts and Applications (4th). Clipping Williams & Wilkins, Philadelphia, 2010.

Examples

## Note that rxode2 parses the names to figure out the best PK parameter

params <- rxDerived(cl = 29.4, v = 23.4, Vp = 114, vp2 = 4614, q = 270, q2 = 73)

## That is why this gives the same results as the value before

params <- rxDerived(CL = 29.4, V1 = 23.4, V2 = 114, V3 = 4614, Q2 = 270, Q3 = 73)

## You may also use micro-constants alpha/beta etc.

params <- rxDerived(k12 = 0.1, k21 = 0.2, k13 = 0.3, k31 = 0.4, kel = 10, v = 10)

## or you can mix vectors and scalars

params <- rxDerived(CL = 29.4, V = 1:3)

## If you want, you can round to a number of significant digits
## with the `digits` argument:

params <- rxDerived(CL = 29.4, V = 1:3, digits = 2)

Jacobian and parameter derivatives

Description

Return Jacobain and parameter derivatives

Usage

rxDfdy(obj)

Arguments

Value

A list of the jacobian parameters defined in this rxode2 object.

Author(s)

Matthew L. Fidler

See Also

Other Query model information: rxInits(), rxLhs(), rxModelVars(), rxParams(), rxState()

Return the DLL associated with the rxode2 object

Description

This will return the dynamic load library or shared object used to run the C code for rxode2.

Usage

rxDll(obj)

Arguments

Value

a path of the library

Author(s)

Matthew L.Fidler

Load rxode2 object

Description

Load rxode2 object

Usage

rxDynLoad(obj)

rxLoad(obj)

Arguments

Value

Boolean returning if the rxode2 library is loaded.

Author(s)

Matthew L.Fidler

Unload rxode2 object

Description

Unload rxode2 object

Usage

rxDynUnload(obj)

rxUnload(obj)

Arguments

Value

Boolean returning if the rxode2 library is loaded.

Author(s)

Matthew L.Fidler

Combine transformations and error structures

Description

Combine error information to figure out what transformation is being applied for the current endpoint

Usage

rxErrTypeCombine(oldErrType, newErrType)

Arguments

Value

The new transformation as a factor

Author(s)

Matthew Fidler

Examples

rxErrTypeCombine("probitNorm")

rxErrTypeCombine("probitNorm") %>%
  rxErrTypeCombine("boxCox")

Dispatch solve to 'rxode2' solve

Description

Dispatch solve to 'rxode2' solve

Usage

rxEtDispatchSolve(x, ...)

## Default S3 method:
rxEtDispatchSolve(x, ...)

Arguments

Value

if 'rxode2' is loaded, a solved object, otherwise an error

Author(s)

Matthew L. Fidler

EVID formatting for tibble and other places.

Description

This is to make an EVID more readable by non pharmacometricians. It displays what each means and allows it to be displayed in a tibble.

Usage

rxEvid(x)

as.rxEvid(x)

## S3 method for class 'rxEvid'
c(x, ...)

## S3 method for class 'rxEvid'
x[...]

## S3 method for class 'rxEvid'
as.character(x, ...)

## S3 method for class 'rxEvid'
x[[...]]

## S3 replacement method for class 'rxEvid'
units(x) <- value

## S3 method for class 'rxRateDur'
c(x, ...)

## S3 method for class 'rxEvid'
format(x, ...)

## S3 method for class 'rxRateDur'
format(x, ...)

## S3 method for class 'rxEvid'
print(x, ...)

Arguments

Value

rxEvid specification

Examples

rxEvid(1:7)

Expand d(f)/d(eta)

Description

Expand d(f)/d(eta)

Usage

rxExpandFEta_(state, neta, pred, isTheta = FALSE)

Arguments

Value

String of symengine expressions to evaluate to calculate df/deta

Faster expand.grid

Description

Only support x and y as characters right now

Usage

rxExpandGrid(x, y, type = 0L)

Arguments

Value

Expand grid

Author(s)

Matthew Fidler

Examples

##
rxExpandGrid(letters, letters)

## Another fast method; See
## https://stackoverflow.com/questions/10405637/use-outer-instead-of-expand-grid

expand.grid.jc <- function(seq1, seq2) {
  cbind(
    Var1 = rep.int(seq1, length(seq2)),
    Var2 = rep.int(seq2, rep.int(length(seq1), length(seq2)))
  )
}

microbenchmark::microbenchmark(rxExpandGrid(letters, letters), expand.grid.jc(letters, letters))

Expand grid internal function

Description

Expand grid internal function

Usage

rxExpandGrid_(c1, c2, type)

Arguments

Value

data frame (when type = 0) or symengine string (when type=1)

Expand if/else clauses into multiple different types of lines.

Description

Expand if/else clauses into multiple different types of lines.

Usage

rxExpandIfElse(model, removeInis = TRUE, removePrint = TRUE)

Arguments

Value

A named character vector. The names of the vector are the logical conditions, the values are the lines that satisfy the logical conditions.

Author(s)

Matthew L. Fidler

Expand second order sensitivity

Description

Expand second order sensitivity

Usage

rxExpandSens2_(state, s1, s2)

Arguments

Value

string for symengine second order sensitivity

Expand sensitivity

Description

Expand sensitivity

Usage

rxExpandSens_(state, calcSens)

Arguments

Value

symengine string for expanded sensitivity

Apply the fixed population estimated parameters

Description

Apply the fixed population estimated parameters

Usage

rxFixPop(ui, returnNull = FALSE)

Arguments

Value

when returnNull is TRUE, NULL if nothing was changed, or the changed model ui. When returnNull is FALSE, return a ui no matter if it is changed or not.

Author(s)

Matthew L. Fidler

Examples

One.comp.transit.allo <- function() {
 ini({
   # Where initial conditions/variables are specified
   lktr <- log(1.15)  #log k transit (/h)
   lcl  <- log(0.15)  #log Cl (L/hr)
   lv   <- log(7)     #log V (L)
   ALLC <- fix(0.75)  #allometric exponent cl
   ALLV <- fix(1.00)  #allometric exponent v
   prop.err <- 0.15   #proportional error (SD/mean)
   add.err <- 0.6     #additive error (mg/L)
   eta.ktr ~ 0.5
   eta.cl ~ 0.1
   eta.v ~ 0.1
 })
 model({
   #Allometric scaling on weight
   cl <- exp(lcl + eta.cl + ALLC * logWT70)
   v  <- exp(lv + eta.v + ALLV * logWT70)
   ktr <- exp(lktr + eta.ktr)
   # RxODE-style differential equations are supported
   d/dt(depot)   = -ktr * depot
   d/dt(central) =  ktr * trans - (cl/v) * central
   d/dt(trans)   =  ktr * depot - ktr * trans
   ## Concentration is calculated
   cp = central/v
   # And is assumed to follow proportional and additive error
   cp ~ prop(prop.err) + add(add.err)
 })
}

m <- rxFixPop(One.comp.transit.allo)
m

# now everything is already fixed, so calling again will do nothing

rxFixPop(m)

# if you call it with returnNull=TRUE when no changes have been
# performed, the function will return NULL

rxFixPop(m, returnNull=TRUE)

Literally fix residual parameters

Description

Literally fix residual parameters

Usage

rxFixRes(ui, returnNull = FALSE)

Arguments

Value

model with residual parameters literally fixed in the model

Author(s)

Matthew L. Fidler

Examples

One.comp.transit.allo <- function() {
 ini({
   # Where initial conditions/variables are specified
   lktr <- log(1.15)  #log k transit (/h)
   lcl  <- log(0.15)  #log Cl (L/hr)
   lv   <- log(7)     #log V (L)
   ALLC <- 0.75  #allometric exponent cl
   ALLV <- 1.00  #allometric exponent v
   prop.err <- fix(0.15)   #proportional error (SD/mean)
   add.err <- fix(0.6)     #additive error (mg/L)
   eta.ktr ~ 0.5
   eta.cl ~ 0.1
   eta.v ~ 0.1
 })
 model({
   #Allometric scaling on weight
   cl <- exp(lcl + eta.cl + ALLC * logWT70)
   v  <- exp(lv + eta.v + ALLV * logWT70)
   ktr <- exp(lktr + eta.ktr)
   # RxODE-style differential equations are supported
   d/dt(depot)   = -ktr * depot
   d/dt(central) =  ktr * trans - (cl/v) * central
   d/dt(trans)   =  ktr * depot - ktr * trans
   ## Concentration is calculated
   cp = central/v
   # And is assumed to follow proportional and additive error
   cp ~ prop(prop.err) + add(add.err)
 })
}

m <- rxFixRes(One.comp.transit.allo)

Clear memoise cache for rxode2

Description

Clear memoise cache for rxode2

Usage

rxForget()

Value

nothing; called for side effects

Author(s)

Matthew L. Fidler

Add/Create C functions for use in rxode2

Description

Add/Create C functions for use in rxode2

Usage

rxFun(name, args, cCode)

rxRmFun(name)

Arguments

Examples

# Right now rxode2 is not aware of the function fun
# Therefore it cannot translate it to symengine or
# Compile a model with it.

try(rxode2("a=fun(a,b,c)"))

# Note for this approach to work, it cannot interfere with C
# function names or reserved rxode2 special terms.  Therefore
# f(x) would not work since f is an alias for bioavailability.

fun <- "
double fun(double a, double b, double c) {
  return a*a+b*a+c;
}
" # C-code for function

rxFun("fun", c("a", "b", "c"), fun) ## Added function

# Now rxode2 knows how to translate this function to symengine

rxToSE("fun(a,b,c)")

# And will take a central difference when calculating derivatives

rxFromSE("Derivative(fun(a,b,c),a)")

## Of course, you could specify the derivative table manually
rxD("fun", list(
  function(a, b, c) {
    paste0("2*", a, "+", b)
  },
  function(a, b, c) {
    return(a)
  },
  function(a, b, c) {
    return("0.0")
  }
))

rxFromSE("Derivative(fun(a,b,c),a)")

# You can also remove the functions by `rxRmFun`

rxRmFun("fun")

# you can also use R functions directly in rxode2


gg <- function(x, y) {
  x + y
}

f <- rxode2({
 z = gg(x, y)
})


e <- et(1:10) %>% as.data.frame()

e$x <- 1:10
e$y <- 21:30

rxSolve(f, e)

# Note that since it touches R, it can only run single-threaded.
# There are also requirements for the function:
#
# 1. It accepts one value per argument (numeric)
#
# 2. It returns one numeric value

# If it is a simple function (like gg) you can also convert it to C
# using rxFun and load it into rxode2

rxFun(gg)

rxSolve(f, e)

# to stop the recompile simply reassign the function
f <- rxode2(f)

rxSolve(f, e)

rxRmFun("gg")
rm(gg)
rm(f)


# You can also automatically convert a R function to R code (and
# calculate first derivatives)

fun <- function(a, b, c) {
  a^2+b*a+c
}

rxFun(fun)

# You can see the R code if you want with rxC

message(rxC("fun"))

# you can also remove both the function and the
# derivatives with rxRmFun("fun")

rxRmFun("fun")

Add user function to rxode2

Description

This adds a user function to rxode2 that can be called. If needed, these functions can be differentiated by numerical differences or by adding the derivatives to rxode2's internal derivative table with rxode2's rxD function

Usage

rxFunParse(name, args, cCode)

rxRmFunParse(name)

Arguments

Value

nothing

Author(s)

Matthew L. Fidler

rxGetControl option from ui

Description

rxGetControl option from ui

Usage

rxGetControl(ui, option, default)

Arguments

Value

Option (if present) or default value

Author(s)

Matthew L. Fidler

This is a S3 method for getting the distribution lines for a rxode2 simulation

Description

This is a S3 method for getting the distribution lines for a rxode2 simulation

Usage

rxGetDistributionSimulationLines(line)

## S3 method for class 'norm'
rxGetDistributionSimulationLines(line)

## S3 method for class 'dnorm'
rxGetDistributionSimulationLines(line)

## S3 method for class 't'
rxGetDistributionSimulationLines(line)

## S3 method for class 'cauchy'
rxGetDistributionSimulationLines(line)

## S3 method for class 'ordinal'
rxGetDistributionSimulationLines(line)

## Default S3 method:
rxGetDistributionSimulationLines(line)

## S3 method for class 'rxUi'
rxGetDistributionSimulationLines(line)

Arguments

Value

Lines for the simulation of ipred and dv. This is based on the idea that the focei parameters are defined

Author(s)

Matthew Fidler

Get the linear compartment model true function

Description

Get the linear compartment model true function

Usage

rxGetLin(model, linCmtSens = c("linCmtA", "linCmtB"), verbose = FALSE)

Arguments

Value

model with linCmt() replaced with linCmtA()

Author(s)

Matthew Fidler

Get model properties without compiling it.

Description

Get model properties without compiling it.

Usage

rxGetModel(
  model,
  calcSens = NULL,
  calcJac = NULL,
  collapseModel = NULL,
  indLin = FALSE
)

Arguments

Value

rxode2 trans list

Author(s)

Matthew L. Fidler

Get the rxode2 seed

Description

Get the rxode2 seed

Usage

rxGetSeed()

Value

rxode2 seed state or -1 when the seed isn't set

See Also

rxSetSeed, rxWithSeed, rxWithPreserveSeed

Examples

# without setting seed

rxGetSeed()
# Now set the seed
rxSetSeed(42)

rxGetSeed()

rxnorm()

rxGetSeed()

# don't use the rxode2 seed again

rxSetSeed(-1)

rxGetSeed()

rxnorm()

rxGetSeed()

Get rxode2 model from object

Description

Get rxode2 model from object

Usage

rxGetrxode2(obj)

Arguments

Value

rxode2 model

Format rxSolve and related objects as html.

Description

Format rxSolve and related objects as html.

Usage

rxHtml(x, ...)

## S3 method for class 'rxSolve'
rxHtml(x, ...)

Arguments

Value

html code for rxSolve object

Author(s)

Matthew L. Fidler

Set the preferred factoring by state

Description

Set the preferred factoring by state

Usage

rxIndLinState(preferred = NULL)

Arguments

Value

Nothing

Author(s)

Matthew Fidler

This sets the inductive linearization strategy for matrix building

Description

When there is more than one state in a ODE that cannot be separated this specifies how it is incorporated into the matrix exponential.

Usage

rxIndLinStrategy(strategy = c("curState", "split"))

Arguments

Value

Nothing

Author(s)

Matthew L. Fidler

Initial Values and State values for a rxode2 object

Description

Returns the initial values of the rxDll object