Estimates the BRAID index of achievable efficacy (or IAE) for a given response surface
estimateIAE(bpar, levels, limits, lowerLimits = c(0, 0))# S3 method for braidrm
estimateIAE(bpar, levels, limits, lowerLimits = c(0, 0))
# S3 method for braidrmflip
estimateIAE(bpar, levels, limits, lowerLimits = c(0, 0))
# S3 method for default
estimateIAE(bpar, levels, limits, lowerLimits = c(0, 0))
estimateFlippedIAE(bpar, flip, levels, limits, lowerLimits = c(0, 0))
A numeric vector, the same length as levels with the estimated
index of achievable efficacy (IAE) values for each of those levels.
The response surface to be evaluated. Can be a numeric vector
(which will be treated as a standard BRAID parameter vector), a BRAID fit
object of class braidrm, or a flipped BRAID fit object of class
braidrmflip
The effect level or levels at which the index is to be estimated
The upper concentration limits beneath which the IAE is to be estimated. Could be known toxicity thresholds, limits on pharmacokinetic availability, or simply a convenient and consistent reference concentration
By default, the IAE is calculated by comparing the area
of dose space below which a given effect level is not reached with the total
achievable dose space specified by limits. However, in some cases, it is
not desirable to allow the sub-threshold area to become infinitesimally
small. If lowerLimits is included, doses that lie below both lower limits
will always be included in the sub-threshold area, placing an effective upper
bound on possible IAE values
String specifying the direction or directions of the surface's flip. Must be one of "A", "B", or "both".
The index of achievable efficacy is an aggregate measure of combined potency,
and a useful first pass for quantifying the efficacy of a given response
surface. Formally, it is equal to the area of achievable dose pairs divided
by the area of achievable doses below which a desired effect level is not
reached (then passed through a square root to give a more dimensionless
result). If the surface is more potent, the area of sub-threshold achievable
doses is smaller, and the IAE is larger. If the surface is less efficacious,
the doses at which a desired effect is reach will be larger, the
sub-threshold area will increase, and the IAE will decrease. By default, the
IAE can range from 1 to infinity, but upper bounds can be placed by setting
the values in lowerLimits to concentrations above 0. For convenience, the
function takes BRAID parameter vectors, braidrm fit objects, and
braidrmflip flipped BRAID fit objects. However, flipped BRAID response
surface parameters cannot be passed to the function as is, so the function
estimateFlippedIAE is also included specifically for flipped parameter
vectors.
estimateIAE(c(1,1,3,3,0,0,100),c(50,90),c(5,5))
bfit <- braidrm(measure ~ concA + concB, synergisticExample, getCIs = FALSE)
estimateIAE(bfit, c(0,0.25,0.5,0.75,1), c(10,1), lowerLimits=c(0.01,0.01))
Run the code above in your browser using DataLab