Package 'pharmr'

Description

Add an admid column to the model dataset and datainfo. Dependent on the presence of a CMT column in order to add admid correctly.

When generated, admids of events in between doses is set to the last used admid.

Usage

add_admid(model)

Arguments

Value

(model : Model) Pharmpy model

See Also

get_admid : Get or create an admid column

get_cmt : Get or create a cmt column

Description

Add allometric scaling of parameters

Add an allometric function to each listed parameter. The function will be P=P*(X/Z)**T where P is the parameter, X the allometric_variable, Z the reference_value and T is a theta. Default is to automatically use clearance and volume parameters.

If there already exists a covariate effect (or allometric scaling) on a parameter with the specified allometric variable, nothing will be added.

If no allometric variable is specified, it will be extracted from the dataset based on the descriptor "body weight".

Usage

add_allometry(
  model,
  allometric_variable = NULL,
  reference_value = 70,
  parameters = NULL,
  initials = NULL,
  lower_bounds = NULL,
  upper_bounds = NULL,
  fixed = TRUE
)

Arguments

Value

(Model) Pharmpy model object

Examples

## Not run: 
model <- load_example_model("pheno")
model <- remove_covariate_effect(model, 'CL', 'WGT')
model <- remove_covariate_effect(model, 'V', 'WGT')
model <- add_allometry(model, allometric_variable='WGT')
model$statements$before_odes

## End(Not run)

Description

Add bioavailability statement for the first dose compartment of the model. Can be added as a new parameter or otherwise it will be set to 1. If added as a parameter, a logit transformation can also be applied.

Usage

add_bioavailability(model, add_parameter = TRUE, logit_transform = FALSE)

Arguments

Value

(Model) Pharmpy model object

See Also

remove_bioavailability

Examples

## Not run: 
model <- load_example_model("pheno")
model <- add_bioavailability(model)

## End(Not run)

Description

Add a CMT column to the model dataset and datainfo if not existed

In case of multiple doses, this method is dependent on the presence of an admid column to correctly number each dose.

NOTE : Existing CMT is based on datainfo type being set to 'compartment' and a column named 'CMT' can be replaced

Usage

add_cmt(model)

Arguments

Value

(model : Model) Pharmpy model

See Also

get_admid : Get or create an admid column

get_cmt : Get or create a cmt column

Description

Adds covariate effect to :class:pharmpy.model.

The following effects have templates:

• Linear function for continuous covariates (lin)

• Function:

(equation could not be rendered, see API doc on website)

• Init: 0.001

• Upper:

• If median of covariate equals minimum: 100,000

• Otherwise: (equation could not be rendered, see API doc on website)

• Lower:

• If median of covariate equals maximum: -100,000

• Otherwise: (equation could not be rendered, see API doc on website)

• Linear function for categorical covariates (cat)

• Function:

• If covariate is the most common category:

(equation could not be rendered, see API doc on website)

• For each additional category:

(equation could not be rendered, see API doc on website)

• Init: 0.001

• Upper: 5

• Lower: -1

• (alternative) Linear function for categorical covariates (cat2)

• Function:

• If covariate is the most common category:

(equation could not be rendered, see API doc on website)

• For each additional category:

(equation could not be rendered, see API doc on website)

• Init: 0.001

• Upper: 6

• Lower: 0

• Piecewise linear function/"hockey-stick", continuous covariates only (piece_lin)

• Function:

• If cov <= median:

(equation could not be rendered, see API doc on website)

• If cov > median:

(equation could not be rendered, see API doc on website)

• Init: 0.001

• Upper:

• For first state: (equation could not be rendered, see API doc on website)

• Otherwise: 100,000

• Lower:

• For first state: -100,000

• Otherwise: (equation could not be rendered, see API doc on website)

• Exponential function, continuous covariates only (exp)

• Function:

(equation could not be rendered, see API doc on website)

• Init:

• If lower > 0.001 or upper < 0.001: (equation could not be rendered, see API doc on website)

• If estimated init is 0: (equation could not be rendered, see API doc on website)

• Otherwise: 0.001

• Upper:

• If min - median = 0 or max - median = 0: 100

• Otherwise:

(equation could not be rendered, see API doc on website)

• Lower:

• If min - median = 0 or max - median = 0: 0.01

• Otherwise:

(equation could not be rendered, see API doc on website)

• Power function, continuous covariates only (pow)

• Function:

(equation could not be rendered, see API doc on website)

• Init: 0.001

• Upper: 100,000

• Lower: -100

Usage

add_covariate_effect(
  model,
  parameter,
  covariate,
  effect,
  operation = "*",
  allow_nested = FALSE
)

Arguments

Value

(Model) Pharmpy model object

Examples

## Not run: 
model <- load_example_model("pheno")
model <- add_covariate_effect(model, "CL", "APGR", "exp")
model$statements$before_odes$full_expression("CL")

## End(Not run)

Description

Add a derivative to be calculcated when running the model. Currently, only derivatives with respect to the prediction is supported. Default is to add all possible ETA and EPS derivatives. First order derivates are specied either by single string or single-element tuple. For instance with_respect_to = "ETA_1" or with_respect_to = ("ETA_1",)

Second order derivatives are specified by giving the two independent varibles in a tuple of tuples. For instance with_respect_to ((ETA_1, EPS_1),)

Multiple derivatives can be specified within a tuple. For instance ((ETA_1, EPS_1), "ETA_1")

Currently, only ETAs and EPSILONs are supported

Usage

add_derivative(model, with_respect_to = NULL)

Arguments

Value

(Pharmpy model.)

Description

Add an effect compartment.

Implemented PD models are:

• Linear:

(equation could not be rendered, see API doc on website)

• Emax:

(equation could not be rendered, see API doc on website)

• Step effect:

(equation could not be rendered, see API doc on website)

• Sigmoidal:

(equation could not be rendered, see API doc on website)

• Log-linear:

(equation could not be rendered, see API doc on website)

(equation could not be rendered, see API doc on website)

Usage

add_effect_compartment(model, expr)

Arguments

Value

(Model) Pharmpy model object

Examples

## Not run: 
model <- load_example_model("pheno")
model <- add_effect_compartment(model, "linear")
model$statements$ode_system$find_compartment("EFFECT")

## End(Not run)

Description

Add estimation step

Adds estimation step for a model in a given index. Methods currently supported are: FO, FOCE, ITS, LAPLACE, IMPMAP, IMP, SAEM

Usage

add_estimation_step(model, method, idx = NULL, ...)

Arguments

Value

(Model) Pharmpy model object

See Also

set_estimation_step

remove_estimation_step

append_estimation_step_options

add_parameter_uncertainty_step

remove_parameter_uncertainty_step

set_evaluation_step

Examples

## Not run: 
model <- load_example_model("pheno")
opts <- list('NITER'=1000, 'ISAMPLE'=100)
model <- add_estimation_step(model, 'IMP', tool_options=opts)
ests <- model$execution_steps
length(ests)
ests[2]

## End(Not run)

Description

Adds IIVs to :class:pharmpy.model.

Effects that currently have templates are:

• Additive (add)

• Proportional (prop)

• Exponential (exp)

• Logit (log)

• Rescaled logit (re_log)

For all except exponential the operation input is not needed. Otherwise user specified input is supported. Initial estimates for new etas are 0.09.

Assuming a statement (equation could not be rendered, see API doc on website)

• Additive: (equation could not be rendered, see API doc on website)

• Proportional: (equation could not be rendered, see API doc on website)

• Exponential: (equation could not be rendered, see API doc on website)

• Logit: (equation could not be rendered, see API doc on website)

• Rescaled logit: (equation could not be rendered, see API doc on website) with (equation could not be rendered, see API doc on website)

Usage

add_iiv(
  model,
  list_of_parameters,
  expression,
  operation = "*",
  initial_estimate = 0.09,
  eta_names = NULL
)

Arguments

Value

(Model) Pharmpy model object

See Also

add_pk_iiv

add_iov

remove_iiv

remove_iov

Examples

## Not run: 
model <- load_example_model("pheno")
model <- remove_iiv(model, "CL")
model <- add_iiv(model, "CL", "add")
model$statements$find_assignment("CL")

## End(Not run)

Description

Add indirect (turnover) effect

The concentration (equation could not be rendered, see API doc on website)

• Production:

(equation could not be rendered, see API doc on website)

• Degradation:

(equation could not be rendered, see API doc on website)

(equation could not be rendered, see API doc on website) Baseline (equation could not be rendered, see API doc on website)

Models:

• Linear:

(equation could not be rendered, see API doc on website)

• Emax:

(equation could not be rendered, see API doc on website)

• Sigmoidal:

(equation could not be rendered, see API doc on website)

Usage

add_indirect_effect(model, expr, prod = TRUE)

Arguments

Value

(Model) Pharmpy model object

Examples

## Not run: 
model <- load_example_model("pheno")
model <- add_indirect_effect(model, expr='linear', prod=TRUE)

## End(Not run)

Description

Add an individual or pk parameter to a model

Usage

add_individual_parameter(model, name)

Arguments

Value

(Model) Pharmpy model object

Examples

## Not run: 
model <- load_example_model("pheno")
model <- add_individual_parameter(model, "KA")
model$statements$find_assignment("KA")

## End(Not run)

Description

Adds IOVs to :class:pharmpy.model.

Initial estimate of new IOVs are 10% of the IIV eta it is based on.

Usage

add_iov(
  model,
  occ,
  list_of_parameters = NULL,
  eta_names = NULL,
  distribution = "disjoint"
)

Arguments

Value

(Model) Pharmpy model object

See Also

add_iiv

add_pk_iiv

remove_iiv

remove_iov

Examples

## Not run: 
model <- load_example_model("pheno")
model <- add_iov(model, "TIME", "CL")
model$statements$find_assignment("CL")

## End(Not run)

Description

Add lag time to the dose compartment of model.

Initial estimate for lag time is set the previous lag time if available, otherwise it is set to the time of first observation/2.

Usage

add_lag_time(model)

Arguments

Value

(Model) Pharmpy model object

See Also

set_transit_compartments

remove_lag_time

Examples

## Not run: 
model <- load_example_model("pheno")
model <- add_lag_time(model)

## End(Not run)

Description

Adds a metabolite compartment to a model

The flow from the central compartment to the metabolite compartment will be unidirectional.

Presystemic indicate that the metabolite compartment will be directly connected to the DEPOT. If a depot compartment is not present, one will be created.

Usage

add_metabolite(model, drug_dvid = 1, presystemic = FALSE)

Arguments

Value

(Model) Pharmpy model object

Examples

## Not run: 
model <- load_example_model("pheno")
model <- add_metabolite(model)

## End(Not run)

Description

Adds parameter uncertainty step to the final estimation step

Usage

add_parameter_uncertainty_step(model, parameter_uncertainty_method)

Arguments

Value

(Model) Pharmpy model object

See Also

add_estimation_step

set_estimation_step

remove_estimation_step

append_estimation_step_options

remove_parameter_uncertainty_step

set_evaluation_step

Examples

## Not run: 
model <- load_example_model("pheno")
model <- set_estimation_step(model, 'FOCE', parameter_uncertainty_method=NULL)
model <- add_parameter_uncertainty_step(model, 'SANDWICH')
ests <- model$execution_steps
ests[1]

## End(Not run)

Description

Adds IIVs to all PD parameters in :class:pharmpy.model.

Usage

add_pd_iiv(model, initial_estimate = 0.09)

Arguments

Value

(Model) Pharmpy model object

See Also

add_iiv

add_iov

remove_iiv

remove_iov

Examples

## Not run: 
model <- load_example_model("pheno")
model <- set_direct_effect(model, 'emax')
model$statements$find_assignment("EC_50")
model <- add_pd_iiv(model)
model$statements$find_assignment("EC_50")

## End(Not run)

Description

Add a peripheral distribution compartment to model

The rate of flow from the central to the peripheral compartment will be parameterized as QPn / VC where VC is the volume of the central compartment. The rate of flow from the peripheral to the central compartment will be parameterized as QPn / VPn where VPn is the volumne of the added peripheral compartment.

If name is set, the peripheral compartment will be added to the compartment with the specified name instead.

Initial estimates:

== =================================================== n == =================================================== 1 (equation could not be rendered, see API doc on website) 2 (equation could not be rendered, see API doc on website) == ===================================================

Usage

add_peripheral_compartment(model, name = NULL)

Arguments

Value

(Model) Pharmpy model object

See Also

set_peripheral_compartment

remove_peripheral_compartment

Examples

## Not run: 
model <- load_example_model("pheno")
model <- add_peripheral_compartment(model)
model$statements$ode_system

## End(Not run)

Description

Adds IIVs to all PK parameters in :class:pharmpy.model.

Will add exponential IIVs to all parameters that are included in the ODE.

Usage

add_pk_iiv(model, initial_estimate = 0.09)

Arguments

Value

(Model) Pharmpy model object

See Also

add_iiv

add_iov

remove_iiv

remove_iov

Examples

## Not run: 
model <- load_example_model("pheno")
model <- set_first_order_absorption(model)
model$statements$find_assignment("MAT")
model <- add_pk_iiv(model)
model$statements$find_assignment("MAT")

## End(Not run)

Description

Add a new population parameter to the model

Usage

add_population_parameter(
  model,
  name,
  init,
  lower = NULL,
  upper = NULL,
  fix = FALSE
)

Arguments

Value

(Model) Pharmpy model object

Examples

## Not run: 
model <- load_example_model("pheno")
model <- add_population_parameter(model, 'POP_KA', 2)
model$parameters

## End(Not run)

Description

Add predictions and/or residuals

Add predictions to estimation step.

Usage

add_predictions(model, pred)

Arguments

Value

(Model) Pharmpy model object

See Also

remove_predictions

remove_residuals

set_estimation_step

add_estimation_step

remove_estimation_step

append_estimation_step_options

add_parameter_uncertainty_step

remove_parameter_uncertainty_step

Examples

## Not run: 
model <- load_example_model("pheno")
model$execution_steps[-1].predictions
model <- add_predictions(model, c('IPRED'))
model$execution_steps[-1].predictions

## End(Not run)

Description

Add predictions and/or residuals

Add residuals to estimation step.

Added redidual variable(s) need to be one of the following : c('RES', 'IRES', 'WRES', 'IWRES', 'CWRES')

Usage

add_residuals(model, res)

Arguments

Value

(Model) Pharmpy model object

See Also

remove_predictions

remove_residuals

set_estimation_step

add_estimation_step

remove_estimation_step

append_estimation_step_options

add_parameter_uncertainty_step

remove_parameter_uncertainty_step

Examples

## Not run: 
model <- load_example_model("pheno")
model$execution_steps[-1].residuals
model <- add_residuals(model, c('WRES'))
model$execution_steps[-1].residuals

## End(Not run)

Description

Calculate and add a TAD column to the dataset

Usage

add_time_after_dose(model)

Arguments

Value

(Model) Pharmpy model object

Examples

## Not run: 
model <- load_example_model("pheno")
model <- add_time_after_dose(model)

## End(Not run)

Description

Append estimation step options

Appends options to an existing estimation step.

Usage

append_estimation_step_options(model, tool_options, idx)

Arguments

Value

(Model) Pharmpy model object

See Also

add_estimation_step

set_estimation_step

remove_estimation_step

add_parameter_uncertainty_step

remove_parameter_uncertainty_step

set_evaluation_step

Examples

## Not run: 
model <- load_example_model("pheno")
opts <- list('NITER'=1000, 'ISAMPLE'=100)
model <- append_estimation_step_options(model, tool_options=opts, idx=0)
est <- model$execution_steps[1]
length(est$tool_options)

## End(Not run)

Description

Bin all observations on the independent variable

Available binning methods:

+—————+————————————————-+ | Method | Description | +===============+=================================================+ | equal_width | Bins with equal width based on the idv | +—————+————————————————-+ | equal_number | Bins containing an equal number of observations | +—————+————————————————-+

Usage

bin_observations(model, method, nbins)

Arguments

Value

(data.frame) A series of bin ids indexed on the original record index of the dataset vector A vector of bin edges

Examples

## Not run: 
model <- load_example_model("pheno")
bins, boundaries <- bin_observations(model, method="equal_width", nbins=10)
bins
boundaries

## End(Not run)

Description

If the model name ends in a number increase it

If path is set increase the number until no file exists with the same name in path. If model name does not end in a number do nothing.

Usage

bump_model_number(model, path = NULL)

Arguments

Value

(Model) Pharmpy model object

Examples

## Not run: 
model <- load_example_model("pheno")
model <- model$replace(name="run2")
model <- bump_model_number(model)
model$name

## End(Not run)

Description

Calculate AIC

AIC = -2LL + 2*n_estimated_parameters

Usage

calculate_aic(model, likelihood)

Arguments

Value

(numeric) AIC of model fit

Description

Calculate BIC

Different variations of the BIC can be calculated:

• | mixed (default) | BIC = -2LL + n_random_parameters * log(n_individuals) + | n_fixed_parameters * log(n_observations)

• | fixed | BIC = -2LL + n_estimated_parameters * log(n_observations)

• | random | BIC = -2LL + n_estimated_parameters * log(n_individuals)

• | iiv | BIC = -2LL + n_estimated_iiv_omega_parameters * log(n_individuals)

Usage

calculate_bic(model, likelihood, type = "mixed")

Arguments

Value

(numeric) BIC of model fit

Examples

## Not run: 
model <- load_example_model("pheno")
results <- load_example_modelfit_results("pheno")
ofv <- results$ofv
calculate_bic(model, ofv)
calculate_bic(model, ofv, type='fixed')
calculate_bic(model, ofv, type='random')
calculate_bic(model, ofv, type='iiv')

## End(Not run)

Description

Calculate correlation matrix from a covariance matrix

Usage

calculate_corr_from_cov(cov)

Arguments

Value

(data.frame) Correlation matrix

See Also

calculate_se_from_cov : Standard errors from covariance matrix

calculate_se_from_prec : Standard errors from precision matrix

calculate_cov_from_prec : Covariance matrix from precision matrix

calculate_cov_from_corrse : Covariance matrix from correlation matrix and standard errors

calculate_prec_from_cov : Precision matrix from covariance matrix

calculate_prec_from_corrse : Precision matrix from correlation matrix and standard errors

calculate_corr_from_prec : Correlation matrix from precision matrix

Examples

## Not run: 
results <- load_example_modelfit_results("pheno")
cov <- results$covariance_matrix
cov
calculate_corr_from_cov(cov)

## End(Not run)

Description

Calculate correlation matrix from a precision matrix

Usage

calculate_corr_from_prec(precision_matrix)

Arguments

Value

(data.frame) Correlation matrix

See Also

calculate_se_from_cov : Standard errors from covariance matrix

calculate_se_from_prec : Standard errors from precision matrix

calculate_corr_from_cov : Correlation matrix from covariance matrix

calculate_cov_from_prec : Covariance matrix from precision matrix

calculate_cov_from_corrse : Covariance matrix from correlation matrix and standard errors

calculate_prec_from_cov : Precision matrix from covariance matrix

calculate_prec_from_corrse : Precision matrix from correlation matrix and standard errors

Examples

## Not run: 
results <- load_example_modelfit_results("pheno")
prec <- results$precision_matrix
prec
calculate_corr_from_prec(prec)

## End(Not run)

Description

Calculate covariance matrix from a correlation matrix and standard errors

Usage

calculate_cov_from_corrse(corr, se)

Arguments

Value

(data.frame) Covariance matrix

See Also

calculate_se_from_cov : Standard errors from covariance matrix

calculate_se_from_prec : Standard errors from precision matrix

calculate_corr_from_cov : Correlation matrix from covariance matrix

calculate_cov_from_prec : Covariance matrix from precision matrix

calculate_prec_from_cov : Precision matrix from covariance matrix

calculate_prec_from_corrse : Precision matrix from correlation matrix and standard errors

calculate_corr_from_prec : Correlation matrix from precision matrix

Examples

## Not run: 
results <- load_example_modelfit_results("pheno")
corr <- results$correlation_matrix
se <- results$standard_errors
corr
calculate_cov_from_corrse(corr, se)

## End(Not run)

Description

Calculate covariance matrix from a precision matrix

Usage

calculate_cov_from_prec(precision_matrix)

Arguments

Value

(data.frame) Covariance matrix

See Also

calculate_se_from_cov : Standard errors from covariance matrix

calculate_se_from_prec : Standard errors from precision matrix

calculate_corr_from_cov : Correlation matrix from covariance matrix

calculate_cov_from_corrse : Covariance matrix from correlation matrix and standard errors

calculate_prec_from_cov : Precision matrix from covariance matrix

calculate_prec_from_corrse : Precision matrix from correlation matrix and standard errors

calculate_corr_from_prec : Correlation matrix from precision matrix

Examples

## Not run: 
results <- load_example_modelfit_results("pheno")
prec <- results$precision_matrix
prec
calculate_cov_from_prec(prec)

## End(Not run)

Description

Calculate the symbolic expression for the epsilon gradient

This function currently only support models without ODE systems

Usage

calculate_epsilon_gradient_expression(model)

Arguments

Value

(Expression) Symbolic expression

See Also

calculate_eta_gradient_expression : Eta gradient

Examples

## Not run: 
model <- load_example_model("pheno_linear")
calculate_epsilon_gradient_expression(model)

## End(Not run)

Description

Calculate the symbolic expression for the eta gradient

This function currently only support models without ODE systems

Usage

calculate_eta_gradient_expression(model)

Arguments

Value

(Expression) Symbolic expression

See Also

calculate_epsilon_gradient_expression : Epsilon gradient

Examples

## Not run: 
model <- load_example_model("pheno_linear")
calculate_eta_gradient_expression(model)

## End(Not run)

Description

Calculate eta shrinkage for each eta

Usage

calculate_eta_shrinkage(
  model,
  parameter_estimates,
  individual_estimates,
  sd = FALSE
)

Arguments

Value

(Series) Shrinkage for each eta

See Also

calculate_individual_shrinkage

Examples

## Not run: 
model <- load_example_model("pheno")
results <- load_example_modelfit_results("pheno")
pe <- results$parameter_estimates
ie <- results$individual_estimates
calculate_eta_shrinkage(model, pe, ie)
calculate_eta_shrinkage(model, pe, ie, sd=TRUE)

## End(Not run)

Description

Calculate statistics for individual parameters

Calculate the mean (expected value of the distribution), variance (variance of the distribution) and standard error for individual parameters described by arbitrary expressions. Any dataset column or variable used in the model can be used in the expression. The exception being that variables that depends on the solution of the ODE system cannot be used. If covariates are used in the expression the statistics of the parameter is calculated at the median value of each covariate as well as at the 5:th and 95:th percentiles. If no parameter uncertainty is available for the model the standard error will not be calculated.

Usage

calculate_individual_parameter_statistics(
  model,
  expr_or_exprs,
  parameter_estimates,
  covariance_matrix = NULL,
  seed = NULL
)

Arguments

Value

(data.frame) A DataFrame of statistics indexed on parameter and covariate value.

Examples

## Not run: 
model <- load_example_model("pheno")
results <- load_example_modelfit_results("pheno")
rng <- create_rng(23)
pe <- results$parameter_estimates
cov <- results$covariance_matrix
calculate_individual_parameter_statistics(model, "K=CL/V", pe, cov, seed=rng)

## End(Not run)

Description

Calculate the individual eta-shrinkage

Definition: ieta_shr = (var(eta) / omega)

Usage

calculate_individual_shrinkage(
  model,
  parameter_estimates,
  individual_estimates_covariance
)

Arguments

Value

(DataFrame) Shrinkage for each eta and individual

See Also

calculate_eta_shrinkage

Examples

## Not run: 
model <- load_example_model("pheno")
results <- load_example_modelfit_results("pheno")
pe <- results$parameter_estimates
covs <- results$individual_estimates_covariance
calculate_individual_shrinkage(model, pe, covs)

## End(Not run)

Description

Scale parameter values from ucp to normal scale

Usage

calculate_parameters_from_ucp(model, scale, ucps)

Arguments

Value

(data.frame) Parameters on the normal scale

See Also

calculate_ucp_scale : Calculate the scale for conversion from ucps

Examples

## Not run: 
model <- load_example_model("pheno")
scale <- calculate_ucp_scale(model)
values <- {'POP_CL': 0.1, 'POP_VC': 0.1, 'COVAPGR': 0.1, 'IIV_CL': 0.1, 'IIV_VC': 0.1, 'SIGMA': 0.1}
calculate_parameters_from_ucp(model, scale, values)

## End(Not run)

Description

Calculate statistics for common pharmacokinetic parameters

Calculate the mean (expected value of the distribution), variance (variance of the distribution) and standard error for some individual pre-defined pharmacokinetic parameters.

Usage

calculate_pk_parameters_statistics(
  model,
  parameter_estimates,
  covariance_matrix = NULL,
  seed = NULL
)

Arguments

Value

(data.frame) A DataFrame of statistics indexed on parameter and covariate value.

See Also

calculate_individual_parameter_statistics : Calculation of statistics for arbitrary parameters

Examples

## Not run: 
model <- load_example_model("pheno")
results <- load_example_modelfit_results("pheno")
rng <- create_rng(23)
pe <- results$parameter_estimates
cov <- results$covariance_matrix
calculate_pk_parameters_statistics(model, pe, cov, seed=rng)

## End(Not run)

Description

Calculate precision matrix from a correlation matrix and standard errors

Usage

calculate_prec_from_corrse(corr, se)

Arguments

Value

(data.frame) Precision matrix

See Also

calculate_se_from_cov : Standard errors from covariance matrix

calculate_se_from_prec : Standard errors from precision matrix

calculate_corr_from_cov : Correlation matrix from covariance matrix

calculate_cov_from_prec : Covariance matrix from precision matrix

calculate_cov_from_corrse : Covariance matrix from correlation matrix and standard errors

calculate_prec_from_cov : Precision matrix from covariance matrix

calculate_corr_from_prec : Correlation matrix from precision matrix

Examples

## Not run: 
results <- load_example_modelfit_results("pheno")
corr <- results$correlation_matrix
se <- results$standard_errors
corr
calculate_prec_from_corrse(corr, se)

## End(Not run)

Description

Calculate precision matrix from a covariance matrix

Usage

calculate_prec_from_cov(cov)

Arguments

Value

(data.frame) Precision matrix

See Also

calculate_se_from_cov : Standard errors from covariance matrix

calculate_se_from_prec : Standard errors from precision matrix

calculate_corr_from_cov : Correlation matrix from covariance matrix

calculate_cov_from_prec : Covariance matrix from precision matrix

calculate_cov_from_corrse : Covariance matrix from correlation matrix and standard errors

calculate_prec_from_corrse : Precision matrix from correlation matrix and standard errors

calculate_corr_from_prec : Correlation matrix from precision matrix

Examples

## Not run: 
results <- load_example_modelfit_results("pheno")
cov <- results$covariance_matrix
cov
calculate_prec_from_cov(cov)

## End(Not run)

Description

Calculate standard errors from a covariance matrix

Usage

calculate_se_from_cov(cov)

Arguments

Value

(data.frame) Standard errors

See Also

calculate_se_from_prec : Standard errors from precision matrix

calculate_corr_from_cov : Correlation matrix from covariance matrix

calculate_cov_from_prec : Covariance matrix from precision matrix

calculate_cov_from_corrse : Covariance matrix from correlation matrix and standard errors

calculate_prec_from_cov : Precision matrix from covariance matrix

calculate_prec_from_corrse : Precision matrix from correlation matrix and standard errors

calculate_corr_from_prec : Correlation matrix from precision matrix

Examples

## Not run: 
results <- load_example_modelfit_results("pheno")
cov <- results$covariance_matrix
cov
calculate_se_from_cov(cov)

## End(Not run)

Description

Calculate standard errors from a precision matrix

Usage

calculate_se_from_prec(precision_matrix)

Arguments

Value

(data.frame) Standard errors

See Also

calculate_se_from_cov : Standard errors from covariance matrix

calculate_corr_from_cov : Correlation matrix from covariance matrix

calculate_cov_from_prec : Covariance matrix from precision matrix

calculate_cov_from_corrse : Covariance matrix from correlation matrix and standard errors

calculate_prec_from_cov : Precision matrix from covariance matrix

calculate_prec_from_corrse : Precision matrix from correlation matrix and standard errors

calculate_corr_from_prec : Correlation matrix from precision matrix

Examples

## Not run: 
results <- load_example_modelfit_results("pheno")
prec <- results$precision_matrix
prec
calculate_se_from_prec(prec)

## End(Not run)

Description

Calculate a scale for unconstrained parameters for a model

The UCPScale object can be used to calculate unconstrained parameters back into the normal parameter space.

Usage

calculate_ucp_scale(model)

Arguments

Value

(UCPScale) A scale object

See Also

calculate_parameters_from_ucp : Calculate parameters from ucp:s

Examples

## Not run: 
model <- load_example_model("pheno")
scale <- calculate_ucp_scale(model)

## End(Not run)

Description

Check dataset for consistency across a set of rules

Usage

check_dataset(model, dataframe = FALSE, verbose = FALSE)

Arguments

Value

(data.frame) Only returns a DataFrame is dataframe=TRUE

Description

Check for highly correlated parameter estimates

Usage

check_high_correlations(model, cor, limit = 0.9)

Arguments

Value

(data.frame) Correlation values indexed on pairs of parameters for (absolute) correlations above limit

Examples

## Not run: 
model <- load_example_model("pheno")
results <- load_example_modelfit_results("pheno")
cor <- results$correlation_matrix
check_high_correlations(model, cor, limit=0.3)

## End(Not run)

Description

Check if any estimated parameter value is close to its bounds

Usage

check_parameters_near_bounds(
  model,
  values,
  zero_limit = 0.001,
  significant_digits = 2
)

Arguments

Value

(data.frame) Logical Series with same index as values

Examples

## Not run: 
model <- load_example_model("pheno")
results <- load_example_modelfit_results("pheno")
check_parameters_near_bounds(model, results$parameter_estimates)

## End(Not run)

Description

Checks whether Pharmpy and pharmr has the same version

Usage

check_pharmpy(pharmpy_version)

Arguments

Description

Perform various cleanups of a model

This is what is currently done

• Make model statements declarative, i.e. only one assignment per symbol

• Inline all assignments of one symbol, e.g. X = Y

• Remove all random variables with no variability (i.e. with omegas fixed to zero)

• Put fixed thetas directly in the model statements

Usage

cleanup_model(model)

Arguments

Value

(Model) Updated model

Note

When creating NONMEM code from the cleaned model Pharmpy might need toadd certain assignments to make it in line with what NONMEM requires.

Examples

## Not run: 
model <- load_example_model("pheno")
model$statements
model <- cleanup_model(model)
model$statements

## End(Not run)

Description

Convert model to other format

Note that the operation is not done inplace.

Usage

convert_model(model, to_format)

Arguments

Value

(Model) New model object with new underlying model format

Examples

## Not run: 
model <- load_example_model("pheno")
converted_model <- convert_model(model, "nlmixr")

## End(Not run)

Description

Creates a basic pk model of given type. The model will be a one compartment model, with first order elimination and in the case of oral administration first order absorption with no absorption delay. The elimination rate will be (equation could not be rendered, see API doc on website)

Usage

create_basic_pk_model(
  administration = "iv",
  dataset_path = NULL,
  cl_init = 0.01,
  vc_init = 1,
  mat_init = 0.1
)

Arguments

Value

(Model) Pharmpy model object

Examples

## Not run: 
model <- create_basic_pk_model('oral')

## End(Not run)

Description

Create a basic config file template

If a configuration file already exists it will not be overwritten

Usage

create_config_template()

Examples

## Not run: 
create_config_template()

## End(Not run)

Description

Combines some or all etas into a joint distribution.

The etas must be IIVs and cannot be fixed. Initial estimates for covariance between the etas is dependent on whether the model has results from a previous run. In that case, the correlation will be calculated from individual estimates, otherwise correlation will be set to 10%.

Usage

create_joint_distribution(model, rvs = NULL, individual_estimates = NULL)

Arguments

Value

(Model) Pharmpy model object

See Also

split_joint_distribution : split etas into separate distributions

Examples

## Not run: 
model <- load_example_model("pheno")
model$random_variables$etas
model <- create_joint_distribution(model, c('ETA_CL', 'ETA_VC'))
model$random_variables$etas

## End(Not run)

Description

Create standard report for results

The report will be an html created at specified path.

Usage

create_report(results, path)

Arguments

Description

Create a new random number generator

Pharmpy functions that use random sampling take a random number generator or seed as input. This function can be used to create a default new random number generator.

Usage

create_rng(seed = NULL)

Arguments

Value

(Generator) Initialized numpy random number generator object

Examples

## Not run: 
rng <- create_rng(23)
rng$standard_normal()

## End(Not run)

Description

Create a new unique variable symbol given a model

Usage

create_symbol(model, stem, force_numbering = FALSE)

Arguments

Value

(Symbol) Created symbol with unique name

Examples

## Not run: 
model <- load_example_model("pheno")
create_symbol(model, "TEMP")
create_symbol(model, "TEMP", force_numbering=TRUE)
create_symbol(model, "CL")

## End(Not run)

Description

Deidentify a dataset

Two operations are performed on the dataset:

1. All ID numbers are randomized from the range 1 to n

2. All columns containing dates will have the year changed

The year change is done by letting the earliest year in the dataset be used as a reference and by maintaining leap years. The reference year will either be 1901, 1902, 1903 or 1904 depending on its distance to the closest preceeding leap year.

Usage

deidentify_data(df, id_column = "ID", date_columns = NULL)

Arguments

Value

(data.frame) Deidentified dataset

Description

Displays the ordinary differential equation system

Usage

display_odes(model)

Arguments

Value

(ODEDisplayer) A displayable object

Examples

## Not run: 
model <- load_example_model("pheno")
display_odes(model)

## End(Not run)

Description

Drop columns from the dataset or mark as dropped

Usage

drop_columns(model, column_names, mark = FALSE)

Arguments

Value

(Model) Pharmpy model object

See Also

drop_dropped_columns : Drop all columns marked as drop

undrop_columns : Undrop columns of model

Examples

## Not run: 
model <- load_example_model("pheno")
model <- drop_columns(model, c('WGT', 'APGR'))
vector(model$dataset$columns)

## End(Not run)

Description

Drop columns marked as dropped from the dataset

NM-TRAN date columns will not be dropped by this function even if marked as dropped. Columns not specified in the datainfo ($INPUT for NONMEM) will also be dropped from the dataset.

Usage

drop_dropped_columns(model)

Arguments

Value

(Model) Pharmpy model object

See Also

drop_columns : Drop specific columns or mark them as drop

Examples

## Not run: 
model <- load_example_model("pheno")
model <- drop_dropped_columns(model)
vector(model$dataset$columns)

## End(Not run)

Description

Evaluate the numeric epsilon gradient

The gradient is evaluated at the current model parameter values or optionally at the given parameter values. The gradient is done for each data record in the model dataset or optionally using the dataset argument. The gradient is done at the current eta values or optionally at the given eta values.

This function currently only support models without ODE systems

Usage

evaluate_epsilon_gradient(
  model,
  etas = NULL,
  parameters = NULL,
  dataset = NULL
)

Arguments

Value

(data.frame) Gradient

See Also

evaluate_eta_gradient : Evaluate the eta gradient

Examples

## Not run: 
model <- load_example_model("pheno_linear")
results <- load_example_modelfit_results("pheno_linear")
etas <- results$individual_estimates
evaluate_epsilon_gradient(model, etas=etas)

## End(Not run)

Description

Evaluate the numeric eta gradient

The gradient is evaluated at the current model parameter values or optionally at the given parameter values. The gradient is done for each data record in the model dataset or optionally using the dataset argument. The gradient is done at the current eta values or optionally at the given eta values.

This function currently only support models without ODE systems

Usage

evaluate_eta_gradient(model, etas = NULL, parameters = NULL, dataset = NULL)

Arguments

Value

(data.frame) Gradient

See Also

evaluate_epsilon_gradient : Evaluate the epsilon gradient

Examples

## Not run: 
model <- load_example_model("pheno_linear")
results <- load_example_modelfit_results("pheno_linear")
etas <- results$individual_estimates
evaluate_eta_gradient(model, etas=etas)

## End(Not run)

Description

Evaluate expression using model

Calculate the value of expression for each data record. The expression can contain dataset columns, variables in model and population parameters. If the model has parameter estimates these will be used. Initial estimates will be used for non-estimated parameters.

Usage

evaluate_expression(model, expression, parameter_estimates = NULL)

Arguments

Value

(data.frame) A series of one evaluated value for each data record

Examples

## Not run: 
model <- load_example_model("pheno")
results <- load_example_modelfit_results("pheno")
pe <- results$parameter_estimates
evaluate_expression(model, "TVCL*1000", parameter_estimates=pe)

## End(Not run)

Description

Evaluate the numeric individual prediction

The prediction is evaluated at the current model parameter values or optionally at the given parameter values. The evaluation is done for each data record in the model dataset or optionally using the dataset argument. The evaluation is done at the current eta values or optionally at the given eta values.

This function currently only support models without ODE systems

Usage

evaluate_individual_prediction(
  model,
  etas = NULL,
  parameters = NULL,
  dataset = NULL
)

Arguments

Value

(data.frame) Individual predictions

See Also

evaluate_population_prediction : Evaluate the population prediction

Examples

## Not run: 
model <- load_example_model("pheno_linear")
results <- load_example_modelfit_results("pheno_linear")
etas <- results$individual_estimates
evaluate_individual_prediction(model, etas=etas)

## End(Not run)

Description

Evaluate the numeric population prediction

The prediction is evaluated at the current model parameter values or optionally at the given parameter values. The evaluation is done for each data record in the model dataset or optionally using the dataset argument.

This function currently only support models without ODE systems

Usage

evaluate_population_prediction(model, parameters = NULL, dataset = NULL)

Arguments

Value

(data.frame) Population predictions

See Also

evaluate_individual_prediction : Evaluate the individual prediction

Examples

## Not run: 
model <- load_example_model("pheno_linear")
results <- load_example_modelfit_results("pheno_linear")
pe <- results$parameter_estimates
evaluate_population_prediction(model, parameters=list(pe))

## End(Not run)

Description

Evaluate the weighted residuals

The residuals is evaluated at the current model parameter values or optionally at the given parameter values. The residuals is done for each data record in the model dataset or optionally using the dataset argument.

This function currently only support models without ODE systems

Usage

evaluate_weighted_residuals(model, parameters = NULL, dataset = NULL)

Arguments

Value

(data.frame) WRES

Examples

## Not run: 
model <- load_example_model("pheno_linear")
results <- load_example_modelfit_results("pheno_linear")
parameters <- results$parameter_estimates
evaluate_weighted_residuals(model, parameters=list(parameters))

## End(Not run)

Description

Expand additional doses into separate dose records

Usage

expand_additional_doses(model, flag = FALSE)

Arguments

Value

(Model) Pharmpy model object

Description

Filter dataset according to expr and return a model with the filtered dataset.

Example: "DVID == 1" will filter the dataset so that only the rows with DVID = 1 remain.

Usage

filter_dataset(model, expr)

Arguments

Value

(Model) Pharmpy model object

Examples

## Not run: 
model <- load_example_model("pheno")
model$dataset
model <- filter_dataset(model, 'WGT < 1.4')
model$dataset

## End(Not run)

Description

Find clearance parameters in model

Usage

find_clearance_parameters(model)

Arguments

Value

(vector) A vector of clearance parameters

Examples

## Not run: 
model <- load_example_model("pheno")
find_clearance_parameters(model)

## End(Not run)

Description

Find volume parameters in model

Usage

find_volume_parameters(model)

Arguments

Value

(vector) A vector of volume parameters

Examples

## Not run: 
model <- load_example_model("pheno")
find_volume_parameters(model)

## End(Not run)

Description

Fit models.

Usage

fit(model_or_models, esttool = NULL, path = NULL, context = NULL)

Arguments

Value

(ModelfitResults | vector of ModelfitResults) ModelfitResults for the model or models

See Also

run_tool

Examples

## Not run: 
model <- load_example_model("pheno")
results <- fit(model)

## End(Not run)

Description

Fix or unfix parameters

Set fixedness of parameters to specified values

Usage

fix_or_unfix_parameters(model, parameters, strict = TRUE)

Arguments

Value

(Model) Pharmpy model object

See Also

fix_parameters : Fix parameters

unfix_paramaters : Unfixing parameters

fix_paramaters_to : Fixing parameters and setting a new initial estimate in the same

function

unfix_paramaters_to : Unfixing parameters and setting a new initial estimate in the same

function

Examples

## Not run: 
model <- load_example_model("pheno")
model$parameters['POP_CL']
model <- fix_or_unfix_parameters(model, list('POP_CL'=TRUE))
model$parameters['POP_CL']

## End(Not run)

Description

Fix parameters

Fix all listed parameters

Usage

fix_parameters(model, parameter_names, strict = TRUE)

Arguments

Value

(Model) Pharmpy model object

See Also

fix_or_unfix_parameters : Fix or unfix parameters (given boolean)

fix_parameters_to : Fixing and setting parameter initial estimates in the same function

unfix_paramaters : Unfixing parameters

unfix_paramaters_to : Unfixing parameters and setting a new initial estimate in the same

function

Examples

## Not run: 
model <- load_example_model("pheno")
model$parameters['POP_CL']
model <- fix_parameters(model, 'POP_CL')
model$parameters['POP_CL']

## End(Not run)

Description

Fix parameters to

Fix all listed parameters to specified value/values

Usage

fix_parameters_to(model, inits, strict = TRUE)

Arguments

Value

(Model) Pharmpy model object

See Also

fix_parameters : Fix parameters

fix_or_unfix_parameters : Fix or unfix parameters (given boolean)

unfix_paramaters : Unfixing parameters

unfix_paramaters_to : Unfixing parameters and setting a new initial estimate in the same

function

Examples

## Not run: 
model <- load_example_model("pheno")
model$parameters['POP_CL']
model <- fix_parameters_to(model, {'POP_CL': 0.5})
model$parameters['POP_CL']

## End(Not run)

Description

Get the admid from model dataset

If an administration column is present this will be extracted otherwise an admid column will be created based on the admids of the present doses. This is dependent on the presence of a CMT column to be generated correctly.

When generated, admids of events in between doses is set to the last used admid.

Usage

get_admid(model)

Arguments

Value

(data.frame) ADMID

Description

Baselines for each subject.

Baseline is taken to be the first row even if that has a missing value.

Usage

get_baselines(model)

Arguments

Value

(data.frame) Dataset with the baselines

Examples

## Not run: 
model <- load_example_model("pheno")
get_baselines(model)

## End(Not run)

Description

Get bioavailability of doses for all compartments

Usage

get_bioavailability(model)

Arguments

Value

(list) Dictionary from compartment name to bioavailability expression

Description

Get the volume and clearance parameters

Usage

get_central_volume_and_clearance(model)

Arguments

Value

(sympy.Symbol) Volume symbol sympy.Symbol Clearance symbol

Examples

## Not run: 
model <- load_example_model("pheno")
get_central_volume_and_clearance(model)

## End(Not run)

Description

Get the cmt (compartment) column from the model dataset

If a cmt column is present this will be extracted otherwise a cmt column will be created. If created, multiple dose compartments are dependent on the presence of an admid type column, otherwise, dose/non-dose will be considered.

Usage

get_cmt(model)

Arguments

Value

(data.frame) CMT

Description

Create a dataframe with concentration parameters

Note that all values are directly calculated from the dataset

Usage

get_concentration_parameters_from_data(model)

Arguments

Value

(data.frame) Concentration parameters

Examples

## Not run: 
model <- load_example_model("pheno")
get_concentration_parameters_from_data(model)

## End(Not run)

Description

Returns path to the user config path

Usage

get_config_path()

Value

(str or NULL) Path to user config or NULL if file does not exist

Examples

## Not run: 
get_config_path()

## End(Not run)

Description

Return a dataframe with baselines of all covariates for each id.

Baseline is taken to be the first row even if that has a missing value.

Usage

get_covariate_baselines(model)

Arguments

Value

(data.frame) covariate baselines

See Also

get_baselines : baselines for all data columns

Examples

## Not run: 
model <- load_example_model("pheno")
model <- set_covariates(model, c("WGT", "APGR"))
get_covariate_baselines(model)

## End(Not run)

Description

Return a list of all used covariates within a model

The list will have parameter name as key with a connected value as a vector of tuple(s) with (covariate, effect type, operator)

Usage

get_covariate_effects(model)

Arguments

Value

(Dictionary : Dictionary of parameters and connected covariate(s))

Description

Get a DOSEID series from the dataset with an id of each dose period starting from 1

If a a dose and observation exist at the same time point the observation will be counted towards the previous dose.

Usage

get_doseid(model)

Arguments

Value

(data.frame) DOSEIDs

Examples

## Not run: 
model <- load_example_model("pheno")
get_doseid(model)

## End(Not run)

Description

Get a series of all doses

Indexed with ID and TIME

Usage

get_doses(model)

Arguments

Value

(data.frame) doses

Examples

## Not run: 
model <- load_example_model("pheno")
get_doses(model)

## End(Not run)

Description

Get the symbol for a certain dvid or dv and check that it is valid

Usage

get_dv_symbol(model, dv = NULL)

Arguments

Value

(sympy.Symbol) DV symbol

Examples

## Not run: 
model <- load_example_model("pheno")
get_dv_symbol(model, "Y")
get_dv_symbol(model, 1)

## End(Not run)

Description

Get the evid from model dataset

If an event column is present this will be extracted otherwise an evid column will be created.

Usage

get_evid(model)

Arguments

Value

(data.frame) EVID

Description

Retrieve a vector of all subject ids of the dataset

Usage

get_ids(model)

Arguments

Value

(vector) All subject ids

Examples

## Not run: 
model <- load_example_model("pheno")
get_ids(model)

## End(Not run)

Description

Retrieves all individual parameters in a :class:pharmpy.model.

By default all individual parameters will be found even ones having no random effect. The level arguments makes it possible to find only those having any random effect or only those having a certain random effect. Using the dv option will give all individual parameters affecting a certain dv. Note that the DV for PD in a PKPD model often also is affected by the PK parameters.

Usage

get_individual_parameters(model, level = "all", dv = NULL)

Arguments

Value

(vectorc(str)) A vector of the parameter names as strings

See Also

get_pd_parameters

get_pk_parameters

get_rv_parameters

has_random_effect

Examples

## Not run: 
model <- load_example_model("pheno")
get_individual_parameters(model)
get_individual_parameters(model, 'iiv')
get_individual_parameters(model, 'iov')

## End(Not run)

Description

Get the full symbolic expression for the modelled individual prediction

This function currently only support models without ODE systems

Usage

get_individual_prediction_expression(model)

Arguments

Value

(Expression) Symbolic expression

See Also

get_population_prediction_expression : Get full symbolic epression for the population prediction

Examples

## Not run: 
model <- load_example_model("pheno_linear")
get_individual_prediction_expression(model)

## End(Not run)

Description

Get initial conditions for the ode system

Default initial conditions at t=0 for amounts is 0

Usage

get_initial_conditions(model, dosing = FALSE)

Arguments

Value

(list) Initial conditions

Examples

## Not run: 
model <- load_example_model("pheno")
get_initial_conditions(model)
get_initial_conditions(model, dosing=TRUE)

## End(Not run)

Description

Get lag times for all compartments

Usage

get_lag_times(model)

Arguments

Value

(list) Dictionary from compartment name to lag time expression

Description

Get MDVs from dataset

Usage

get_mdv(model)

Arguments

Value

(data.frame) MDVs

Description

Get the model code of the underlying model language

Usage

get_model_code(model)

Arguments

Value

(str) Model code

Examples

## Not run: 
model <- load_example_model("pheno")
get_model_code(model)

## End(Not run)

Description

List of covariates used in model

A covariate in the model is here defined to be a data item affecting the model prediction excluding dosing items that are not used in model code.

Usage

get_model_covariates(model, strings = FALSE)

Arguments

Value

(vector) Covariate symbols or names

Examples

## Not run: 
model <- load_example_model("pheno")
get_model_covariates(model)
get_model_covariates(model, strings=TRUE)

## End(Not run)

Description

Return Mu name connected to parameter

If the given parameter is not dependent on any Mu, NULL is returned

Usage

get_mu_connected_to_parameter(model, parameter)

Arguments

Value

(str) Name of Mu parameter or NULL

Description

Retrieve the number of individuals in the model dataset

Usage

get_number_of_individuals(model)

Arguments

Value

(integer) Number of individuals in the model dataset

Note

For NONMEM models this is the number of individuals of the active dataset, i.e. after filteringof IGNORE and ACCEPT and removal of individuals with no observations.

See Also

get_number_of_observations : Get the number of observations in a dataset

get_number_of_observations_per_individual : Get the number of observations per individual in a

dataset

Examples

## Not run: 
model <- load_example_model("pheno")
get_number_of_individuals(model)

## End(Not run)

Description

Retrieve the total number of observations in the model dataset

Usage

get_number_of_observations(model)

Arguments

Value

(integer) Number of observations in the model dataset

Note

For NONMEM models this is the number of observations of the active dataset, i.e. after filteringof IGNORE and ACCEPT and removal of individuals with no observations.

See Also

get_number_of_individuals : Get the number of individuals in a dataset

get_number_of_observations_per_individual : Get the number of observations per individual in a

dataset

Examples

## Not run: 
model <- load_example_model("pheno")
get_number_of_observations(model)

## End(Not run)

Description

Number of observations for each individual

Usage

get_number_of_observations_per_individual(model)

Arguments

Value

(data.frame) Number of observations in the model dataset

Note

For NONMEM models this is the individuals and number of observations of the active dataset, i.e.after filtering of IGNORE and ACCEPT and removal of individuals with no observations.

See Also

get_number_of_individuals : Get the number of individuals in a dataset

get_number_of_observations_per_individual : Get the number of observations per individual in a

dataset

Examples

## Not run: 
model <- load_example_model("pheno")
get_number_of_observations_per_individual(model)

## End(Not run)

Description

Return the number of peripherals compartments connected to the central compartment

Usage

get_number_of_peripheral_compartments(model)

Arguments

Value

(integer) Number of peripherals compartments

Description

Return the number of transit compartments in the model

Usage

get_number_of_transit_compartments(model)

Arguments

Value

(integer) Number of transit compartments

Description

Get the full symbolic expression for the observation according to the model

This function currently only support models without ODE systems

Usage

get_observation_expression(model)

Arguments

Value

(Expression) Symbolic expression

Examples

## Not run: 
model <- load_example_model("pheno_linear")
expr <- get_observation_expression(model)
print(expr$unicode())

## End(Not run)

Description

Get observations from dataset

Usage

get_observations(model, keep_index = FALSE)

Arguments

Value

(data.frame) Observations indexed over ID and TIME

See Also

get_number_of_observations : get the number of observations

get_number_of_observations_per_individual : get the number of observations per individual

Examples

## Not run: 
model <- load_example_model("pheno")
get_observations(model)

## End(Not run)

Description

Get all omegas (variability parameters) of a model

Usage

get_omegas(model)

Arguments

Value

(Parameters) A copy of all omega parameters

See Also

get_thetas : Get theta parameters

get_sigmas : Get sigma parameters

Examples

## Not run: 
model <- load_example_model("pheno")
get_omegas(model)

## End(Not run)

Description

Retrieves name of random variable in :class:pharmpy.model.Model given a parameter.

Usage

get_parameter_rv(model, parameter, var_type = "iiv")

Arguments

Value

(vectorc(str)) A vector of random variable names for the given parameter

See Also

get_rv_parameters

has_random_effect

get_pk_parameters

get_individual_parameters

Examples

## Not run: 
model <- load_example_model("pheno")
get_parameter_rv(model, 'CL')

## End(Not run)

Description

Retrieves PD parameters in :class:pharmpy.model.Model.

Usage

get_pd_parameters(model)

Arguments

Value

(vectorc(str)) A vector of the PD parameter names of the given model

See Also

get_pk_parameters

Examples

## Not run: 
model <- load_example_model("pheno")
model <- set_direct_effect(model, "linear")
get_pd_parameters(model)

## End(Not run)

Description

Retrieves PK parameters in :class:pharmpy.model.Model.

Usage

get_pk_parameters(model, kind = "all")

Arguments

Value

(vectorc(str)) A vector of the PK parameter names of the given model

See Also

get_individual_parameters

get_rv_parameters

Examples

## Not run: 
model <- load_example_model("pheno")
get_pk_parameters(model)
get_pk_parameters(model, 'absorption')
get_pk_parameters(model, 'distribution')
get_pk_parameters(model, 'elimination')

## End(Not run)

Description

Get the full symbolic expression for the modelled population prediction

This function currently only support models without ODE systems

Usage

get_population_prediction_expression(model)

Arguments

Value

(Expression) Symbolic expression

See Also

get_individual_prediction_expression : Get full symbolic epression for the individual prediction

Examples

## Not run: 
model <- load_example_model("pheno_linear")
get_population_prediction_expression(model)

## End(Not run)

Description

Retrieves parameters in :class:pharmpy.model.Model given a random variable.

Usage

get_rv_parameters(model, rv)

Arguments

Value

(vectorc(str)) A vector of parameter names for the given random variable

See Also

has_random_effect

get_pk_parameters

get_individual_parameters

Examples

## Not run: 
model <- load_example_model("pheno")
get_rv_parameters(model, 'ETA_CL')

## End(Not run)

Description

Get all sigmas (residual error variability parameters) of a model

Usage

get_sigmas(model)

Arguments

Value

(Parameters) A copy of all sigma parameters

See Also

get_thetas : Get theta parameters

get_omegas : Get omega parameters

Examples

## Not run: 
model <- load_example_model("pheno")
get_sigmas(model)

## End(Not run)

Description

Get all thetas (structural parameters) of a model

Usage

get_thetas(model)

Arguments

Value

(Parameters) A copy of all theta parameters

See Also

get_omegas : Get omega parameters

get_sigmas : Get sigma parameters

Examples

## Not run: 
model <- load_example_model("pheno")
get_thetas(model)

## End(Not run)

Description

Derive the physical unit of a variable in the model

Unit information for the dataset needs to be available. The variable can be defined in the code, a dataset olumn, a parameter or a random variable.

Usage

get_unit_of(model, variable)

Arguments

Value

(Unit) A unit expression

Examples

## Not run: 
model <- load_example_model("pheno")
get_unit_of(model, "Y")
get_unit_of(model, "VC")
get_unit_of(model, "WGT")

## End(Not run)

Description

Get zero order inputs for all compartments

Usage

get_zero_order_inputs(model)

Arguments

Value

(sympy.Matrix) Vector of inputs

Examples

## Not run: 
model <- load_example_model("pheno")
get_zero_order_inputs(model)

## End(Not run)

Description

Convert to using greek letters for all population parameters

Usage

greekify_model(model, named_subscripts = FALSE)

Arguments

Value

(Model) Pharmpy model object

Examples

## Not run: 
model <- load_example_model("pheno")
model$statements
model <- greekify_model(cleanup_model(model))
model$statements

## End(Not run)

Description

Check if a model has an additive error model

Multiple dependent variables are supported. By default the only (in case of one) or the first (in case of many) dependent variable is going to be checked.

Usage

has_additive_error_model(model, dv = NULL)

Arguments

Value

(logical) TRUE if the model has an additive error model and FALSE otherwise

See Also

has_proportional_error_model : Check if a model has a proportional error model

has_combined_error_model : Check if a model has a combined error model

has_weighted_error_model : Check if a model has a weighted error model

Examples

## Not run: 
model <- load_example_model("pheno")
has_additive_error_model(model)

## End(Not run)

Description

Check if a model has a combined additive and proportional error model

Multiple dependent variables are supported. By default the only (in case of one) or the first (in case of many) dependent variable is going to be checked.

Usage

has_combined_error_model(model, dv = NULL)

Arguments

Value

(logical) TRUE if the model has a combined error model and FALSE otherwise

See Also

has_additive_error_model : Check if a model has an additive error model

has_proportional_error_model : Check if a model has a proportional error model

has_weighted_error_model : Check if a model has a weighted error model

Examples

## Not run: 
model <- load_example_model("pheno")
has_combined_error_model(model)

## End(Not run)

Description

Tests if an instance of :class:pharmpy.model has a given covariate effect.

Usage

has_covariate_effect(model, parameter, covariate)

Arguments

Value

(logical) Whether input model has a covariate effect of the input covariate on the input parameter.

Examples

## Not run: 
model <- load_example_model("pheno")
has_covariate_effect(model, "CL", "APGR")

## End(Not run)

Description

Check if ode system describes a first order absorption

Currently defined as the central compartment having a unidirectional input flow from another compartment (such as depot or transit)

Usage

has_first_order_absorption(model)

Arguments

Value

(Bool : TRUE if model has first order absorption)

Description

Check if the model describes first order elimination

This function relies on heuristics and will not be able to detect all possible ways of coding the first order elimination.

Usage

has_first_order_elimination(model)

Arguments

Value

(logical) TRUE if model has describes first order elimination

Examples

## Not run: 
model <- load_example_model("pheno")
has_first_order_elimination(model)

## End(Not run)

Description

Check if ode system describes a instantaneous absorption

Defined as being a instantaneous dose directly into the central compartment

Usage

has_instantaneous_absorption(model)

Arguments

Value

(Bool : TRUE if model has instantaneous absorption)

Description

Check if model has a linear ODE system

Usage

has_linear_odes(model)

Arguments

Value

(logical) TRUE if model has an ODE system that is linear

See Also

has_odes

has_linear_odes_with_real_eigenvalues

Examples

## Not run: 
model <- load_example_model("pheno")
has_linear_odes(model)

## End(Not run)

Description

Check if model has a linear ode system with real eigenvalues

Usage

has_linear_odes_with_real_eigenvalues(model)

Arguments

Value

(logical) TRUE if model has an ODE system that is linear

See Also

has_odes

has_linear_odes

Examples

## Not run: 
model <- load_example_model("pheno")
has_linear_odes_with_real_eigenvalues(model)

## End(Not run)

Description

Check if the model describes Michaelis-Menten elimination

This function relies on heuristics and will not be able to detect all possible ways of coding the Michaelis-Menten elimination.

Usage

has_michaelis_menten_elimination(model)

Arguments

Value

(logical) TRUE if model has describes Michaelis-Menten elimination

Examples

## Not run: 
model <- load_example_model("pheno")
has_michaelis_menten_elimination(model)
model <- set_michaelis_menten_elimination(model)
has_michaelis_menten_elimination(model)

## End(Not run)

Description

Check if the model describes mixed Michaelis-Menten and first order elimination

This function relies on heuristics and will not be able to detect all possible ways of coding the mixed Michalis-Menten and first order elimination.

Usage

has_mixed_mm_fo_elimination(model)

Arguments

Value

(logical) TRUE if model has describes Michaelis-Menten elimination

Examples

## Not run: 
model <- load_example_model("pheno")
has_mixed_mm_fo_elimination(model)
model <- set_mixed_mm_fo_elimination(model)
has_mixed_mm_fo_elimination(model)

## End(Not run)

Description

Check if model is Mu-reference or not.

Will return TRUE if each parameter with an ETA is dependent on a Mu parameter.

Usage

has_mu_reference(model)

Arguments

Value

(logical) Whether the model is mu referenced

Description

Check if model has an ODE system

Usage

has_odes(model)

Arguments

Value

(logical) TRUE if model has an ODE system

See Also

has_linear_odes

has_linear_odes_with_real_eigenvalues

Examples

## Not run: 
model <- load_example_model("pheno")
has_odes(model)

## End(Not run)

Description

Checks whether a model has a presystemic metabolite

If pre-systemic drug there will be a flow from DEPOT to METABOLITE as well as being a flow from the CENTRAL to METABOLITE

Usage

has_presystemic_metabolite(model)

Arguments

Value

(logical) Whether a model has presystemic metabolite

Examples

## Not run: 
model <- load_example_model("pheno")
model <- add_metabolite(model, presystemic=TRUE)
has_presystemic_metabolite(model)

## End(Not run)

Description

Check if a model has a proportional error model

Multiple dependent variables are supported. By default the only (in case of one) or the first (in case of many) dependent variable is going to be checked.

Usage

has_proportional_error_model(model, dv = NULL)

Arguments

Value

(logical) TRUE if the model has a proportional error model and FALSE otherwise

See Also

has_additive_error_model : Check if a model has an additive error model

has_combined_error_model : Check if a model has a combined error model

has_weighted_error_model : Check if a model has a weighted error model

Examples

## Not run: 
model <- load_example_model("pheno")
has_proportional_error_model(model)

## End(Not run)

Description

Decides whether the given parameter of a :class:pharmpy.model has a random effect.

Usage

has_random_effect(model, parameter, level = "all")

Arguments

Value

(logical) Whether the given parameter has a random effect

See Also

get_individual_parameters

get_rv_parameters

Examples

## Not run: 
model <- load_example_model("pheno")
has_random_effect(model, 'S1')
has_random_effect(model, 'CL', 'iiv')
has_random_effect(model, 'CL', 'iov')

## End(Not run)

Description

Check if ode system describes a sequential zero-order, first-order absorption

Defined as the model having both zero- and first-order absorption.

Usage

has_seq_zo_fo_absorption(model)

Arguments

See Also

has_zero_order_absorption

has_first_order_absorption

Description

Check if a model has a weighted error model

Usage

has_weighted_error_model(model)

Arguments

Value

(logical) TRUE if the model has a weighted error model and FALSE otherwise

See Also

has_additive_error_model : Check if a model has an additive error model

has_combined_error_model : Check if a model has a combined error model

has_proportional_error_model : Check if a model has a proportional error model

Examples

## Not run: 
model <- load_example_model("pheno")
has_weighted_error_model(model)

## End(Not run)

Description

Check if ode system describes a zero order absorption

currently defined as having Infusion dose with rate not in dataset

Usage

has_zero_order_absorption(model)

Arguments

Value

(logical) Whether the model has zero order absorption or not

Examples

## Not run: 
model <- load_example_model("pheno")
has_zero_order_absorption(model)

## End(Not run)

Description

Check if the model describes zero-order elimination

This function relies on heuristics and will not be able to detect all possible ways of coding the zero-order elimination.

Usage

has_zero_order_elimination(model)

Arguments

Value

(logical) TRUE if model has describes zero order elimination

Examples

## Not run: 
model <- load_example_model("pheno")
has_zero_order_elimination(model)
model <- set_zero_order_elimination(model)
has_zero_order_elimination(model)

## End(Not run)

Description

Install the pharmpy-core python package into virtual environment. Uses the same Pharmpy version as pharmr.

Usage

install_pharmpy(envname = "r-reticulate", method = "auto")

Arguments

Description

Install the pharmpy-core python package into virtual environment.

Usage

install_pharmpy_devel(
  envname = "r-reticulate",
  method = "auto",
  version = "devel"
)

Arguments

Description

Determine if a model is linearized

Usage

is_linearized(model)

Arguments

Value

(logical) TRUE if model has been linearized and FALSE otherwise

Examples

## Not run: 
model1 <- load_example_model("pheno")
is_linearized(model1)
model2 <- load_example_model("pheno_linear")
is_linearized(model2)

## End(Not run)

Description

Determine if an expression is real valued given constraints of a model

Usage

is_real(model, expr)

Arguments

Value

(logical or NULL) TRUE if expression is real, FALSE if not and NULL if unknown

Examples

## Not run: 
model <- load_example_model("pheno")
is_real(model, "CL")

## End(Not run)

Description

Takes a ModelfitResults object and a statement as input and returns TRUE/FALSE if the evaluation of the statement is TRUE/FALSE.

Usage

is_strictness_fulfilled(model, results, strictness)

Arguments

Value

(logical) A logical indicating whether the strictness criteria are fulfilled or not.

Examples

## Not run: 
res <- load_example_modelfit_results('pheno')
model <- load_example_model('pheno')
is_strictness_fulfilled(model, res, "minimization_successful or rounding_errors")

## End(Not run)

Description

Return a vector of names of all time varying covariates

Usage

list_time_varying_covariates(model)

Arguments

Value

(vector) Names of all time varying covariates

See Also

get_covariate_baselines : get baselines for all covariates

Examples

## Not run: 
model <- load_example_model("pheno")
list_time_varying_covariates(model)

## End(Not run)

Description

Load the dataset given datainfo

Usage

load_dataset(model)

Arguments

Value

(Model) Pharmpy model with dataset removed

Examples

## Not run: 
model <- load_example_model("pheno")
model <- unload_dataset(model)
model$dataset is NULL
model <- load_dataset(model)
model$dataset

## End(Not run)

Description

Load an example model

Load an example model from models built into Pharmpy

Usage

load_example_model(name)

Arguments

Value

(Model) Loaded model object

Examples

## Not run: 
model <- load_example_model("pheno")
model$statements

## End(Not run)

Description

Load the modelfit results of an example model

Load the modelfit results of an example model built into Pharmpy

Usage

load_example_modelfit_results(name)

Arguments

Value

(ModelfitResults) Loaded modelfit results object

Examples

## Not run: 
results <- load_example_modelfit_results("pheno")
results$parameter_estimates

## End(Not run)

Description

Make the model statments declarative

Each symbol will only be declared once.

Usage

make_declarative(model)

Arguments

Value

(Model) Pharmpy model object

Examples

## Not run: 
model <- load_example_model("pheno")
model$statements$before_odes
model <- make_declarative(model)
model$statements$before_odes

## End(Not run)

Description

Convert model to use mu-referencing

Mu-referencing an eta is to separately define its actual mu (mean) parameter. For example: (equation could not be rendered, see API doc on website) normal distribution would give (equation could not be rendered, see API doc on website) (equation could not be rendered, see API doc on website)

Usage

mu_reference_model(model)

Arguments

Value

(Model) Pharmpy model object

Examples

## Not run: 
model <- load_example_model("pheno")
model <- mu_reference_model(model)
model$statements$before_odes

## End(Not run)

Description

Iterate over omissions of a certain group in a dataset. One group is omitted at a time.

Usage

omit_data(dataset_or_model, group, name_pattern = "omitted_{}")

Arguments

Value

(iterator) Iterator yielding tuples of models/dataframes and the omitted group

Description

Plot \|CWRES\| vs IPRED

Usage

plot_abs_cwres_vs_ipred(
  model,
  predictions,
  residuals,
  stratify_on = NULL,
  bins = 8
)

Arguments

Value

(alt.Chart) Plot

Examples

## Not run: 
model <- load_example_model("pheno")
res <- load_example_modelfit_results("pheno")
plot_abs_cwres_vs_ipred(model, res$predictions, res$residuals)
model <- load_example_model("pheno")
res <- load_example_modelfit_results("pheno")
plot_abs_cwres_vs_ipred(model, res$predictions, res$residuals, 'WGT', bins=4)

## End(Not run)

Description

Plot CWRES vs idv

Usage

plot_cwres_vs_idv(model, residuals, stratify_on = NULL, bins = 8)

Arguments

Value

(alt.Chart) Plot

Examples

## Not run: 
model <- load_example_model("pheno")
res <- load_example_modelfit_results("pheno")
plot_cwres_vs_idv(model, res$residuals)
model <- load_example_model("pheno")
res <- load_example_modelfit_results("pheno")
plot_cwres_vs_idv(model, res$residuals, 'WGT', bins=4)

## End(Not run)

Description

Plot DV vs IPRED

Usage

plot_dv_vs_ipred(model, predictions, stratify_on = NULL, bins = 8)

Arguments

Value

(alt.Chart) Plot

Examples

## Not run: 
model <- load_example_model("pheno")
res <- load_example_modelfit_results("pheno")
plot_dv_vs_ipred(model, res$predictions)
model <- load_example_model("pheno")
res <- load_example_modelfit_results("pheno")
plot_dv_vs_ipred(model, res$predictions, 'WGT', bins=4)

## End(Not run)

Description

Plot DV vs PRED

Usage

plot_dv_vs_pred(model, predictions, stratify_on = NULL, bins = 8)

Arguments

Value

(alt.Chart) Plot

Examples

## Not run: 
model <- load_example_model("pheno")
res <- load_example_modelfit_results("pheno")
plot_dv_vs_pred(model, res$predictions)
model <- load_example_model("pheno")
res <- load_example_modelfit_results("pheno")
plot_dv_vs_pred(model, res$predictions, 'WGT', bins=4)

## End(Not run)

Description

Plot eta distributions for all etas

Usage

plot_eta_distributions(model, individual_estimates)

Arguments

Value

(alt.Chart) Plot

Examples

## Not run: 
model <- load_example_model("pheno")
res <- load_example_modelfit_results("pheno")
plot_eta_distributions(model, res$individual_estimates)

## End(Not run)

Description

Plot DV and predictions grouped on individuals

Usage

plot_individual_predictions(model, predictions, individuals = NULL)

Arguments

Value

(alt.Chart) Plot

Examples

## Not run: 
model <- load_example_model("pheno")
res <- load_example_modelfit_results("pheno")
plot_individual_predictions(model, res$predictions, individuals=c(1, 2, 3, 4, 5))

## End(Not run)