TDR: Time-Dose-Response analysis pipeline

In jcduda/td2pLL: Fitting a time-dose two-parameter log-logistic model to time-dose-response data

Time-Dose-Response analysis pipeline

Description

TDR performs the time-dose-response analysis pipeline as presented in Duda et al. (2021). That is: For a dose-response or concentration-response data set where, additionally also the (exposure) time is varied, this procedure can be applied. The main aim of this procedure are cytotoxicity data. The approach is divided into two steps. In step one, it is tested in a nested ANOVA if the (exposure) time has an influence on the dose-response relationship. In case of a significant results, the time-dose two-parameter log-logistic model is fitted to the data:

f(d,t)=100-100\frac{d^h}{EC_{50}(t)^h + d^h}

with

EC_{50}(t) = Δ \cdot t^{-γ} + C_0.

If no significant result is obtained, a dose-response 2pLL curve is fitted, ignoring the information on (exposure) time.

Usage

TDR(data, alpha = 0.05, strict_stop = FALSE, ...)

Arguments

data	(Numeric data.frame()) Data frame with columns named time, dose and resp. Note that the data is expected to be on the percent scale with values (roughly) within 0 and 100.
alpha	(numeric(1)in (0,1)) 1- alpha is the confidence level for testing in step 1.
strict_stop	(logical(1)) Optional logical. When FALSE, the default, then in case of an error due to non-convergence in the pre-test, then in the second step a simple 2pLL model is fitted as if the pre-test was non- significant. If strict_stop is TRUE and there is an error due to non-convergence in the pre-test, the procedure stops and no model is fitted in step 2.
...	Further arguments that can be passed on to fit_td2pLL().

Details

For further details on the td2pLL model, check fit_td2pLL(). For details on the ANOVA used, see td2pLL_anova(). More over, the entire procedure is explained in duda et al. (2021).

Value

A list with entries pretest and fit. pretest is captures the anova based pre-test result as a list with entires signif (TRUE/FALSE or NA if no-convergence), alpha, anova_res (the anova result from function anova) and conv (logical: If the pre-test converged). fit Is, depending on the pre-test, either an object of class td2pLL or a 2pLL fit, i.e. an object of class drc.

Examples

data(cytotox)
data_subset <- cytotox[cytotox$compound == "ASP", c("expo", "dose", "resp")]
colnames(data_subset)[1] <- "time"
TDR_res <- TDR(data = data_subset)
# Pre-test rejected time dependency, so a regular 2pLL model is the result
plot(TDR_res$fit)
data_subset <- cytotox[cytotox$compound == "CHL", c("expo", "dose", "resp")]
colnames(data_subset)[1] <- "time"
TDR_res <- TDR(data = data_subset)
# Pre-test did not reject time dependency, so a  td2pLL model is the result
# Note that the interactive Plot is in the Viewer panel, not in the Plots panel
plot(TDR_res$fit)

jcduda/td2pLL documentation built on May 14, 2022, 6:48 p.m.