Example (preset)
Choose an example to fill inputs and see results immediately.
- Paste concentration and response (replicate columns supported) or import a CSV/TSV file.
- Select direction (inhibition/activation) and model (auto/4PL/5PL).
- IC50/EC50 (50% point), fitted curve, and fit metrics are shown.
This is an analysis tool. 4PL/5PL definitions can differ across software. Check the equation and the definition of the “50% point”. In 5PL, the inflection parameter may differ from the 50% point.
Input (paste / CSV)
Format: col 1 = concentration, col 2+ = response (rep1, rep2, …). TSV/CSV supported.
Settings (minimum needed)
Results
Plots
Table (mean±SD, predicted, residuals)
| Concentration | mean | SD | Predicted | Residual | Excluded |
|---|
Parameters
| Parameter | Estimate | 95% CI (when enabled) |
|---|
The 95% CI is an approximation (linearization). Do not over-trust it when you have few points or outliers. In 5PL, the 50% point may differ from the inflection-parameter EC50, so interpret results with the notes in the result cards.
4PL/5PL and the 50% point
This tool fits a dose–response curve with 4PL (symmetric) or 5PL (asymmetric) and calculates the midpoint between top and bottom (the 50% point) as IC50/EC50.
Fitting uses Levenberg–Marquardt (nonlinear least squares). With SD weighting enabled, each point is weighted by 1/SD2.
Equations (reference)
Here, concentration x is positive.
- 4PL:
y = bottom + (top-bottom) / (1 + (x/EC50)^hill) - 5PL:
y = bottom + (top-bottom) / (1 + (x/EC50_param)^hill)^asym
In 5PL, EC50_param is not always the “50% point”. This tool computes the 50% point separately and shows it explicitly.
A zero concentration (control) can help for normalization, but it cannot be used on a log x-axis, so excluding it from fitting is usually safer.
How to use this calculator effectively
Use this page to estimate the 50% point from dose-response data, compare 4PL with 5PL, and review fit quality before you export or share results.
1. Prepare the table first
Put concentration in the first column and replicate responses in the following columns. Keep concentration units consistent across all rows, and decide whether zero-concentration controls should be used only for normalization or also shown in the raw table.
2. Choose the simplest model that explains the curve
Start with auto or 4PL. Move to 5PL only when the left and right sides of the sigmoid are clearly asymmetric. A lower AIC can help, but do not rely on a small numerical difference alone.
3. Use point exclusion sparingly
Exclude a point only when you have an experimental reason, such as a failed well or a pipetting error. Check whether the direction, IC50/EC50, and residual pattern remain plausible after exclusion.
4. Interpret the 50% point and pIC50 together
This tool shows the 50% point explicitly. In 5PL, it may differ from the inflection parameter. If the concentration unit can be converted to molar, pIC50/pEC50 helps compare potency across datasets.
See also
FAQ
What is the difference between IC50 and EC50?
What is the difference between 4PL and 5PL?
Is the 5PL EC50 parameter the same as the 50% point?
Can I include a zero concentration (control)?
How are replicates handled?
How is pIC50 calculated?
Related tools
- ELISA standard curve fitter (4PL/5PL) | CalcBEcalculators, elisa-standard-curve, calculator, en
- Kd calculator | Binding curve fit (one site / Hill) | CalcBEcalculators, kd-binding-curve, calculator, en
- Growth curve fitter (exponential & logistic) | CalcBEcalculators, growth-curve, calculator, biology, growth, curve, fitter, exponential, logistic, en
- Protein standard curve calculator (BCA / Bradford) | CalcBEcalculators, protein-standard-curve, calculator, en
- qPCR standard curve calculator | CalcBEcalculators, qpcr-standard-curve, calculator, biology, qpcr, standard, curve, efficiency, r², unknowns, en