sim1sbind: Simulate one-site ligand receptor binding

In TJMurphy/nlfitr: Fitting models to nonlinear data

Description

A sandbox to simulate and visualize random normal heteroscedastic response data. Variances enlarge with the value of y predicted by the model using a constant coefficeint of variation (cv). The data generating formula is derived from the one-site total binding model: y=Bmax*x/(x+kd). Failure errors in the plot fitting subfunction will occasionally happen due to the random data. These are more frequent with higher cv values. Just re-simulate with modified parameter values.

Usage

sim1sbind(x, bmax, kd, cv, reps, log = F)

Arguments

x	a vector representing dose or concentration of ligand (exponential or linear values)
bmax	the measured value where the receptor population is completely saturated by ligand (i.e. maximum binding)
kd	the value of x that yields y/Bmax = 0.5 (the equilibrium binding constant)
cv	the coefficient of variation for y replicates
reps	an integer value for number of replicates
log	logical value. Default is FALSE. If TRUE, linear x values are transformed using a log10 function for plotting. Only for visual aesthetic.

Value

ggplot, data

Examples

dose <- c(1, 3, 10, 30, 100, 300) # eg, in nM units
logdose <- c(1e-3, 3e-3, 1e-2, 3e-2, 1e-1, 3e-1, 1e0, 3e0) # eg, in nM units

set.seed(2345)

binddat <- sim1sbind(dose, bmax = 1000, kd = 50, cv = 0.10, reps = 5, log = FALSE ); binddat

binddat$data #extract the data frame containing x and cv-modified y values

sim1sbind(logdose, bmax = 10000, kd = 5e-2, cv = 0.20, reps = 5, log = TRUE)

TJMurphy/nlfitr documentation built on March 18, 2021, 12:33 p.m.