Perform visual predictive check for NLME models

Usage

vpcmodel(
  model,
  vpcParams,
  params,
  hostPlatform = NULL,
  runInBackground = FALSE,
  ...
)

Arguments

model: PK/PD model class object.
vpcParams: VPC argument setup. See NlmeVpcParams. If missing, default values generated by NlmeVpcParams() are used.
params: Engine argument setup. See engineParams. The following arguments are the subject of interest: sort, ODE, rtolODE, atolODE, maxStepsODE. If missing, default values generated by engineParams(model) are used.
hostPlatform: Host definition for model execution. See hostParams. If missing, simple local host is used.
runInBackground: Set to TRUE to run in background and return prompt.
...: Additional class initializer arguments for NlmeVpcParams or hostParams, or arguments available inside engineParams functions. If engineParams arguments are supplied through both params argument and additional argument (i.e., ellipsis), then the arguments in params will be ignored and only the additional arguments will be used with warning. If hostParams arguments are supplied through both hostPlatform argument and additional argument, then its values will be overridden by additional arguments. In addition, if NlmeVpcParams arguments are supplied through both vpcParams argument and additional argument, then its slots will be overridden by additional arguments.

Value

if runInBackground is TRUE, it returns job properties. Otherwise,

If the function is called in an interactive mode, the resulting simulated tables and summary statistics tables will be loaded and presented as a list;
If the function is called in a non-interactive mode, it returns the full paths of the tables generated

Examples

if (FALSE) { # \dontrun{
model <- pkmodel(
  numComp = 1,
  absorption = "Extravascular",
  ID = "Subject",
  Time = "Act_Time",
  CObs = "Conc",
  Aa = "Amount",
  data = pkData,
  modelName = "PkModel",
  workingDir = tempdir()
)

 host <- hostParams(
  sharedDirectory = tempdir(),
  parallelMethod = "NONE",
  hostName = "local",
  numCores = 1
 )

job <- fitmodel(model = model,
                hostPlatform = host)

finalModelVPC <- copyModel(model,
                           acceptAllEffects = TRUE,
                           modelName = "model_VPC",
                           workingDir = tempdir())

# View the model
print(finalModelVPC)

# Set up VPC arguments to have PRED outputted to simulation output dataset "predout.csv"
vpcSetup <- NlmeVpcParams(outputPRED = TRUE)

# Run VPC using the default host, default values for the relevant NLME engine arguments
finalVPCJob <- vpcmodel(model = finalModelVPC, vpcParams = vpcSetup, hostPlatform = host)
# the same as:
# finalVPCJob <- vpcmodel(model = finalModelVPC, outputPRED = TRUE)

# Observed dataset predcheck0.csv
dt_ObsData <- finalVPCJob$predcheck0

# Simulation output dataset predout.csv
dt_SimData <- finalVPCJob$predout

# Add PRED from REPLICATE = 0 of simulation output dataset to observed input dataset
dt_ObsData$PRED <- dt_SimData[REPLICATE == 0]$PRED

# tidyvpc package VPC example:
# library(tidyvpc)
# library(magrittr)
# Create a regular VPC plot with binning method set to be "jenks"
# binned_VPC <- observed(dt_ObsData, x = IVAR, yobs = DV) %>%
# simulated(dt_SimData, ysim = DV) %>%
# binning(bin = "jenks") %>%
# vpcstats()

# plot_binned_VPC <- plot(binned_VPC)

# Create a pcVPC plot with binning method set to be "jenks"
# binned_pcVPC <- observed(dt_ObsData, x = IVAR, yobs = DV) %>%
#   simulated(dt_SimData, ysim = DV) %>%
#   binning(bin = "jenks") %>%
#   predcorrect(pred = PRED) %>%
#   vpcstats()

# plot_binned_pcVPC <- plot(binned_pcVPC)
} # }