PKNCA has many options that control its function. These options have
effects throughout the package. The options are controlled using either
the PKNCA.options
function or by passing the
options
argument to any of the functions with that as an
argument. All options supported by the current version of PKNCA (0.11.0)
are listed below with their descriptions.
The adjusted r^2 for the calculation of lambda.z has this factor times the number of data points added to it. It allows for more data points to be preferred in the calculation of half-life.
The default value is: 1e-04
The maximum fraction of the data that may be missing (‘NA’) to calculate summary statistics with the business.* functions.
The default value is: 0.5
The method used to calculate the AUC and related statistics. Options are: “lin up/log down”, “linear”, “lin-log”
The default value is: lin up/log down
How should missing (‘NA’) concentration values be handled? See help for ‘clean.conc.na’ for how to use this option.
The default value is: drop
How should below the limit of quantification (zero, 0) concentration values be handled? See help for ‘clean.conc.blq’ for how to use this option.
$first [1] “keep”
$middle [1] “drop”
$last [1] “keep”
If there is more than one concentration equal to Cmax, which time should be selected for Tmax? If ‘TRUE’, the first will be selected. If ‘FALSE’, the last will be selected.
The default value is: TRUE
Should the concentration and time at Tmax be allowed in the half-life calculation? ‘TRUE’ is yes and ‘FALSE’ is no.
The default value is: FALSE
What additional columns from the intervals should be kept in the results?
NULL
What is the minimum number of points required to calculate half-life?
The default value is: 3
What is the minimum span ratio required to consider a half-life valid?
The default value is: 2
What is the maximum percent extrapolation to consider an AUCinf valid?
The default value is: 20
What is the minimum r-squared value to consider a half-life calculation valid?
The default value is: 0.9
A value to pass to purrr::pmap(.progress = ) to create a progress bar while running
The default value is: TRUE
What values for tau (repeating interdose interval) should be considered when attempting to automatically determine the intervals for multiple dosing? See ‘choose.auc.intervals’ and ‘find.tau’ for more information. ‘NA’ means automatically look at any potential interval.
The default value is: NA
When data is single-dose, what intervals should be used?
start | end | auclast | aucall | aumclast | aumcall | aucint.last | aucint.last.dose | aucint.all | aucint.all.dose | c0 | cmax | cmin | tmax | tlast | tfirst | clast.obs | cl.last | cl.all | f | mrt.last | mrt.iv.last | vss.last | vss.iv.last | cav | cav.int.last | cav.int.all | ctrough | cstart | ptr | tlag | deg.fluc | swing | ceoi | aucabove.predose.all | aucabove.trough.all | count_conc | totdose | ae | clr.last | clr.obs | clr.pred | fe | sparse_auclast | sparse_auc_se | sparse_auc_df | time_above | aucivlast | aucivall | aucivint.last | aucivint.all | aucivpbextlast | aucivpbextall | aucivpbextint.last | aucivpbextint.all | half.life | r.squared | adj.r.squared | lambda.z | lambda.z.time.first | lambda.z.n.points | clast.pred | span.ratio | thalf.eff.last | thalf.eff.iv.last | kel.last | kel.iv.last | aucinf.obs | aucinf.pred | aumcinf.obs | aumcinf.pred | aucint.inf.obs | aucint.inf.obs.dose | aucint.inf.pred | aucint.inf.pred.dose | aucivinf.obs | aucivinf.pred | aucivpbextinf.obs | aucivpbextinf.pred | aucpext.obs | aucpext.pred | cl.obs | cl.pred | mrt.obs | mrt.pred | mrt.iv.obs | mrt.iv.pred | mrt.md.obs | mrt.md.pred | vz.obs | vz.pred | vss.obs | vss.pred | vss.iv.obs | vss.iv.pred | vss.md.obs | vss.md.pred | cav.int.inf.obs | cav.int.inf.pred | thalf.eff.obs | thalf.eff.pred | thalf.eff.iv.obs | thalf.eff.iv.pred | kel.obs | kel.pred | kel.iv.obs | kel.iv.pred | auclast.dn | aucall.dn | aucinf.obs.dn | aucinf.pred.dn | aumclast.dn | aumcall.dn | aumcinf.obs.dn | aumcinf.pred.dn | cmax.dn | cmin.dn | clast.obs.dn | clast.pred.dn | cav.dn | ctrough.dn |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
0 | 24 | TRUE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE |
0 | Inf | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | TRUE | FALSE | TRUE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | TRUE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | TRUE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE | FALSE |