Dose–Response with 95% Confidence Intervals
A four-parameter logistic drawn over the observed responses on a logarithmic concentration axis, with the observations carrying 95% confidence intervals that say so. Shows the two things a dose–response figure is usually wrong about: an unlabelled uncertainty, and a log axis whose base is never stated.
Make it your own.
title: Inhibition of target kinase
subtitle: Eight concentrations, n = 6 wells per concentration
x: Concentration (nM)
y: Response (% of vehicle control)
x scale: log
y limits: 0 .. 110
grid: y
legend: bottom
scatter "Observed" error: ci
1, 97.2 ± 3.1
3, 94.8 ± 3.4
10, 88.1 ± 4.0
30, 71.4 ± 4.6
100, 47.9 ± 5.1
300, 24.6 ± 4.2
1000, 11.3 ± 3.0
3000, 6.8 ± 2.4
note: The curve below is the fitted 4PL with the reported EC50 of 92 nM and a Hill slope of −1.1. It is drawn from those constants, not refitted here.
plot "4PL fit" y = 5.5 + 92/(1 + (x/92)^1.1) for x in [1, 3000]