Regression & multivariate

IC50 / EC50 / LD50 Calculator: Dose–Response Curves (4PL, Probit and Logit)

Two analyses in one place. For a continuous response (cell viability, enzyme activity, root length, % inhibition), fit a 4- or 3-parameter log-logistic curve and get the IC50/EC50, EC10 and EC90 with 95% confidence intervals. For quantal data (number dead out of number tested), run probit or logit analysis and get the LD50/LC50, LD10 and LD90 with delta-method and Fieller intervals, a goodness-of-fit test and the slope.

✅ No login needed100% freeRuns in your browserYour data stays on your device
Dose–Response and IC50 Calculator✅ Free · no login📚 Cite this tool
Your data never leaves your PC. Your data is analysed on your own computer or phone, even when you upload a CSV or Excel file. Nothing is uploaded, stored or seen by Codanics, and no login is needed.

  • PriceFree, no login or sign-up needed
  • Your dataProcessed in your browser, never uploaded
  • Works onPhone, tablet or computer, any modern browser
  • Made byDr. Muhammad Aammar Tufail, Codanics
  • Published17 November 2026
  • Last updated16 November 2026

📚 How to cite this tool

This tool is free. If it helped your thesis, paper, report or assignment, please cite it. It keeps the tools free and helps other students find them.

Citation · APA 7th edition

Tufail, M. A. (2026). Dose–response and IC50 calculator (4PL, probit and logit) [Online tool]. Codanics. Retrieved 16 November 2026, from https://codanics.com/tools/ic50-calculator/

In-text citation: (Tufail, 2026) · Narrative: Tufail (2026)

The access date is today’s date, filled in automatically. Author: Muhammad Aammar Tufail. Published 17 November 2026.

🚀 Follow Codanics channel on WhatsApp for more updatesNew free tools, scholarships, PhD positions and research tips, straight to your phone. Free lessons on YouTube too.

How to use the Dose–Response and IC50 Calculator

  1. Choose the tab: Continuous response for viability, absorbance, growth or % inhibition (IC50/EC50), or Mortality counts for dead/alive data (LD50/LC50).
  2. Continuous: enter one dose per line followed by its replicate responses. The ryegrass root-length data (drc package) are loaded as an example. Choose whether to estimate the bottom and top plateaus or fix them (for example 0 and 100 when the data are normalised to % of control).
  3. Mortality: enter dose, number tested and number dead on each line, with dose 0 for the control. A simulated insecticide bioassay is loaded as an example. Choose probit or logit and, if controls died, Abbott’s correction.
  4. Click the button to fit. Read the IC50/EC50 or LD50 with its 95% CI, the slope and the goodness of fit, and check the curve.
  5. Copy the tables or the results sentence, download the CSV and save the curve as PNG or SVG.

IC50, EC50, LD50: what they mean

  • IC50 (half-maximal inhibitory concentration): the dose that reduces a response (cell viability, enzyme activity, growth) by half of the full inhibitory effect.
  • EC50 (half-maximal effective concentration): the same for a response that increases with dose.
  • LD50 / LC50 (median lethal dose / concentration): the dose that kills 50% of the test organisms, estimated from counts of dead and alive individuals.

Lower values mean a more potent compound. Always report the value with its 95% confidence interval and the slope.

Relative vs absolute IC50

The relative IC50 is the inflection point e of the fitted curve: halfway between the fitted bottom and top plateaus. The absolute IC50 is the dose at which the response crosses 50% of the control (for example 50% viability). They are equal only when the curve runs from 100 to 0. Use the “Absolute IC at response” box to get the absolute value, or fix the bottom at 0 and the top at 100 for normalised data.

Models

Log-logistic (drc LL.4): y = c + (d − c) / (1 + exp(b·(ln x − ln e))) b slope, c bottom, d top, e = IC50/EC50 ECp (p% of the full effect): x = e · (p/(100 − p))^(1/|b|) SE by the delta method; CI = estimate ± t(df)·SE Probit: Φ⁻¹(P) = a + b·log₁₀(dose) Logit: ln(P/(1 − P)) = a + b·log₁₀(dose) LDp = 10^((z_p − a)/b) Fieller CI for m = (z_p − a)/b: [m − g·v_ab/v_bb ± (t/|b|)·√(v_aa − 2m·v_ab + m²·v_bb − g·(v_aa − v_ab²/v_bb))] / (1 − g), g = t²·v_bb/b² Heterogeneity factor h = χ²/df; if used, Cov × h and t(df) instead of z (Finney, 1971) Abbott: P = (P_obs − C)/(1 − C)

The continuous fits use the same parameterisation as the R package drc (Ritz et al., 2015): on the ryegrass example, the parameters, standard errors and ED10/ED50/ED90 with delta CIs match drm(rootl ~ conc, fct = LL.4()) and ED() to at least five significant digits (standard errors come from the observed information matrix, as in drc). In drc a positive slope b means a decreasing curve. The probit and logit fits match glm(cbind(dead, alive) ~ log10(dose), family = binomial(link = "probit")), and the delta-method intervals match MASS::dose.p().

Designing a good dose–response experiment

  • Use 6–10 doses spaced evenly on a log scale (for example two- or three-fold dilutions), so that both plateaus and the steep middle part are covered.
  • Run at least three replicates per dose and include untreated controls.
  • For mortality bioassays, aim for doses giving between about 10% and 90% mortality, with 20–50 individuals per dose, and keep control mortality below 10% (correct 5–20% with Abbott’s formula).
  • Prepare the dilution series with the serial dilution calculator.
Learn the statistics behind these numbers. Codanics has free statistics, R and Python lessons on YouTube and full courses for researchers.

Frequently asked questions

How do I calculate IC50 from my data?

Enter each dose with its replicate responses (for example % viability from an MTT assay) and click Fit. The calculator fits a four-parameter logistic curve and reports the IC50 (the inflection point) with a 95% confidence interval. Doses can be in any unit; do not log-transform them yourself, the tool does that.

Should I fix the top and bottom of the curve?

Fix them only when you have good reason, for example data normalised to 100% (untreated) and 0% (fully killed) controls, or when your doses do not reach one plateau. Otherwise let the curve estimate them. A 3PL with bottom = 0 is common when the response can drop to zero.

What is the difference between probit and logit analysis?

Both model mortality on log dose. Probit assumes a normal tolerance distribution and is the classic method (Finney); logit assumes a logistic distribution with slightly heavier tails. They usually give very similar LD50s; LD10 and LD90 differ more. Use probit to compare with older literature.

What is the Fieller confidence interval?

It is the exact interval for a ratio of two estimated quantities, here the LD50 on the log scale. It is asymmetric and widens strongly when the slope is poorly determined. It is the standard interval in probit analysis software such as POLO and SPSS Probit.

What is Abbott's correction?

If some control individuals die, part of the mortality at each dose is natural. Abbott’s formula, corrected mortality = (observed − control)/(1 − control), removes it. It is usually applied when control mortality is between 5% and 20%; with more than 20% the test should be repeated.

Why is my IC50 outside the doses I tested?

Because the curve did not reach 50% effect within your doses, so the value is extrapolated and unreliable. Extend the dose range, or report it as ‘greater than’ the highest dose tested.

Does it match GraphPad Prism?

The 4PL fit is the same model as Prism’s ‘log(inhibitor) vs response, variable slope (four parameters)’, written with a natural-log dose. The IC50 and its standard error agree; Prism’s default CI is profile-likelihood based in recent versions, so CIs can differ slightly from the delta-method CIs here.

Do I need to log in or create an account to use the Dose–Response and IC50 Calculator?

No. There is no login, no sign-up and no email required. Open the page and start using it straight away, for free, on any device.

How do I cite this tool in my thesis or paper?

Use the citation box on this page. In APA 7th edition: Tufail, M. A. (2026). Dose–response and IC50 calculator (4PL, probit and logit) [Online tool]. Codanics. Retrieved [date you used it], from https://codanics.com/tools/ic50-calculator/. BibTeX, RIS (EndNote, Mendeley, Zotero), MLA, Harvard, Chicago and Vancouver formats are also available.

Free guides and courses from Codanics

More free tools

See all free Codanics tools → · Codanics on YouTube · WhatsApp channel

Made by Dr. Muhammad Aammar Tufail, Codanics. First published 17 November 2026, last updated 16 November 2026. Found a bug or have an idea for a new tool? Tell us on the WhatsApp channel or in the comments on YouTube. Results are for learning and planning; check important analyses with a statistician or your supervisor.