Simulate TDM collection — collect

Using the "ground truth" model, simulate collection of drug/biomarker levels. Returns a data frame of sample time, true value, and measured value (with residual error added).

Usage

collect_tdms(
  sim_model,
  t_obs,
  res_var,
  pars_i,
  lloq = NULL,
  est_model = NULL,
  est_pars_i = NULL,
  ...
)

Arguments

sim_model: model used for simulated patient response ("truth").
t_obs: sample collection times (since start of treatment)
res_var: data frame of residual variability quantities to add. See pregenerate_ruv for details. Expects columns prop and add. Error is added in the linear domain, so log-transformed models should supply error as proportional error with 0 additive error.
pars_i: true parameters for the individual (named list)
lloq: lower limit of quantification. If non-NULL, all TDMs below LLOQ will be set to half the LLOQ.
est_model: model used for estimation (e.g. the model used in MAP fitting). If provided, a predictive individual prediction is simulated using est_pars_i and stored in the predictive_ipred column.
est_pars_i: parameters for est_model. Typically population or MAP-estimated individual parameters.
...: arguments passed on to PKPDsim::sim

Value

a data frame with columns t (time), true_y (actual level), y (measured level), and predictive_ipred (predicted level from estimation model; NA if est_model is not supplied), with rows corresponding to t_obs.