curveFit: curveFit

In BenjaminChittick/BenCScore: Methods For Analysis of High-Throughput Screens

Description

When the only tool you have is a hammer, every problem looks like a nail. curveFit, fits a 4 parameter logistic inhibition curve according to y[i] ~ N(mu[i], tau) mu <- top + (bottom - top)/(1 + 10^((log(IC50) - log(x))*hillslope))

Usage

curveFit(x, y, top.mu, bottom.mu, runjags.method = "rjparallel")

Arguments

x	Numeric of compound concentrations
y	Numeric measured values aligned with x

Details

IC50 <- exp(1/100) hillslope ~ dunif(0.1, 3) top ~ dnorm(0, 0.01) bottom ~ dnorm(-100, 0.01) tau <- pow(sigma, -2) sigma ~ dunif(0, 100)

Examples

x <- c(2, 4, 8, 16, 32, 64, 128, 256)
y <- -100/(1 + 10^((log(27) - log(x))*1.23))
y.error <- y + rnorm(length(y), 0, 5)
plot(y.error, x)
estimate <- curveFit(x, y.error)
estimate

BenjaminChittick/BenCScore documentation built on May 5, 2019, 2:41 p.m.