doseresponse / medrc Goto Github PK

adding self start functionality to medrc model functions

#' @param x numeric dose vector.
#' @param b steepness
#' @param c lower limit
#' @param d upper limit
#' @param e ED50
meLL.4 <- function(x, b, c, d, e){
  .value <- c + (d-c)/(1 + exp(b*log(x/e)))
  # ...
}

The above is not a conventional way of writing the logistic model that I've seen. Instead I usually see it written as:

fourPL <- function(x, a, b, c, d, e) {
  d + (c - d) / (1 + ((x / e)^b))
}

In the formulation used by this software, curves that exhibit increasing response with respect to dose have negative slope. Is there a reason for this or is it just historical? I understand that changing it now would be a problematic breaking change, but I found it pretty unintuitive.

Store residual variance of individual drm model fits in metadrm()

Transfer data examples to drc data packages

summary() method
predict() method
ranef() method
VarCorr() method

Hi,

I'm working on analyzing dose response data with a clear hormesis effect from a blocked experiment, however, when I set fct = BC.4() I get the following error:

Error in str2lang(x) : :1:18: unexpected ','
1: germination~(time,
^

Here's an example dataset and my model:

#dataset
time <- c(0,0,0,0,12,12,12,12,24,24,24,24,48,48,48,48,96,96,96,96)
block <- c(1,2,3,4,1,2,3,4,1,2,3,4,1,2,3,4,1,2,3,4)
block <- as.factor(block)
germination <- c(.78,.80,.82,.75,.90,.95,.89,.91,.68,.71,.65,.69,.08,.05,.06,.02,0,0,.01,0)
data <- data.frame(block, time, germination)

#model
m <- medrm(germination ~ time,
data = data,
fct = BC.4(),
random = b + e ~ 1|block)

I haven't been able to find any other mentions of str2lang errors from drc/medrc online. Does medrc support the Brain-Cousens model? If not, can you point me in the direction of other functions to use for the model?

Thanks!

Copying drc model functions into medrc that can be directly used in nlme.
e.g. meLL.4(), meW1.4() ...