R/aab_growthmodel-constructor.R

In growthrates: Estimate Growth Rates from Experimental Data

#' Create a User-defined Parametric Growth Model
#'
#' This constructor method allows to create user-defined functions
#'   that can be used as parametric models describing time-dependent
#'   growth of organisms.
#'
#' Package \pkg{growthrates} has a plug-in architecture allowing
#'   user-defined growth models of the following form:
#'
#' \preformatted{
#'   identifier <- function(time, parms) {
#'     ... content of function here ...
#'     return(as.matrix(data.frame(time=time, y=y)))
#'   }
#' }
#'
#' where \code{time} is a numeric vector and \code{parms} a named, non-nested
#'     list of model parameters. The constructor function \code{growthmodel}
#'     is used to attach the names of the parameters as an optional
#'     attribute.
#'
#' @param x a function with arguments \code{times} and \code{parms}, and
#'   returning a matrix with two columns \code{time} and \code{y}.
#' @param pnames character vector with the names of the model parameters.
#'
#' @examples
#'
#' test <- function(time, parms) {
#'   with(as.list(parms), {
#'     y <- (K * y0) / (y0 + (K - y0) * exp(-mumax * time)) + y_shift
#'     return(as.matrix(data.frame(time=time, y=y)))
#'  })
#' }
#'
#' mygrowthmodel <- growthmodel(test, c("y0", "mumax", "K", "y_shift"))
#'
#'
#' @family growth models
#' @rdname growthmodel-constructor
#' @aliases user-defined functions
#' @export growthmodel
#'
growthmodel <- function(x, pnames=NULL) {
  if (!is.function(x)) stop("X must be a function")
  structure(x, pnames=pnames, class = c("growthmodel", "function"))
}