Enter your standard concentrations and corresponding absorbance values. The calculator will fit a linear regression and calculate sample concentrations.
| # | Concentration (pg/mL) | Absorbance (OD) |
|---|
If samples were diluted before the assay, multiply concentration by this factor.
Enter your standard curve data for 4-parameter logistic (4PL) fitting. This is the most accurate method for sigmoid ELISA curves.
| # | Concentration (pg/mL) | Absorbance (OD) |
|---|
📊 ELISA Standard Curve Results
Standard Curve
Sample Concentrations
| Sample | Absorbance (OD) | Concentration | After Dilution |
|---|
Introduction
The ELISA Calculator is a free browser-based tool for researchers and students who need to convert raw absorbance readings into analyte concentrations using a standard curve. It supports both linear regression and 4-parameter logistic (4PL) curve fitting, making it suitable for a wide range of enzyme-linked immunosorbent assay formats used in immunology, clinical diagnostics, and biomarker research.
About the Tool
This ELISA Calculator builds a standard curve from known standards and back-calculates the concentration of unknown samples from their absorbance readings. It is designed for sandwich ELISAs, competitive ELISAs, and any assay format that produces a dose-response relationship between concentration and absorbance signal.
Two fitting tabs are provided: Linear Fit for assays confined to the linear portion of the curve, and 4PL Fit for the sigmoidal (S-shaped) curves typical of sandwich and competitive ELISAs. The tool requires a minimum of 3 standard points for linear fitting and 4 points for 4PL fitting, though 6–8 points are recommended for a well-constrained curve. All calculations run locally in your browser — standard curve data, sample absorbances, and results are never sent to a server.
Input Explanation
- Curve fitting method. Use the tab bar to choose between Linear Fit and 4PL Fit. For most sandwich ELISAs that span a wide concentration range, 4PL is strongly recommended because it accounts for the sigmoidal shape of the curve at high and low concentrations.
- Standard curve data. Enter the concentration and absorbance (OD) for each standard well. Click "Load Example" to see a typical dataset if you are using the tool for the first time.
- Concentration unit. Choose pg/mL, ng/mL, µg/mL, ng/µL, or µM depending on the units used in your standard curve.
- Sample absorbance values. Type your sample absorbance readings separated by commas into the "Sample Absorbance values" field.
- Sample dilution factor. If you diluted your sample before loading it onto the plate, enter the dilution factor (e.g., 10 for a 1:10 dilution). The calculator multiplies the back-calculated concentration by this factor to give the actual concentration in the original undiluted sample.
Formula Explanation
Linear regression fits the standard data to the equation: Absorbance = slope × Concentration + intercept. The concentration of each sample is then calculated by rearranging: Concentration = (Absorbance − intercept) / slope. The goodness of fit is expressed as R², where values ≥ 0.99 are typically required for assay acceptance.
4-Parameter Logistic (4PL) regression fits the standard data to the equation:
Where: A = minimum asymptote (absorbance at zero concentration), B = Hill slope (steepness of the curve), C = EC50 or inflection point (concentration at the midpoint of the curve), D = maximum asymptote (absorbance at saturating concentration). Units for A, B and D are absorbance (OD, dimensionless); C carries the same concentration unit as the standards. To back-calculate concentration from absorbance, the equation is inverted: x = C × ((A − D) / (y − D) − 1)^(1/B).
Worked Example
Sample Input
A sandwich ELISA for a cytokine is run with 8 standards spanning 0–2000 pg/mL (loaded via "Load Example" on the 4PL tab), plus 3 unknown serum samples that were pre-diluted 1:10 before plating. Concentration unit is set to pg/mL, the sample absorbances entered are 0.450, 0.950, 1.600 OD, and the Sample Dilution Factor is set to 10.
Step-by-Step Calculation
Fitting the 4PL model to the 8 standards converges to A = −0.146, B = 1.023, C = 250.0 pg/mL, D = 2.010, with R² = 0.954. For each sample, the inverted 4PL equation x = C × ((A − D) / (y − D) − 1)^(1/B) is applied to the absorbance to interpolate a concentration, and that concentration is then multiplied by the dilution factor of 10 to back-calculate the concentration in the original undiluted serum.
Final Result
| Sample | Absorbance (OD) | Interpolated Conc. | After ×10 Dilution |
|---|---|---|---|
| Sample 1 | 0.450 | 97.597 pg/mL | 975.971 pg/mL |
| Sample 2 | 0.950 | 258.259 pg/mL | 2582.590 pg/mL |
| Sample 3 | 1.600 | 1029.917 pg/mL | 10299.171 pg/mL |
Interpretation
All three sample absorbances fall within the range defined by the standards (0.04–1.98 OD), so none are flagged out of range (OOR) and the interpolated values can be trusted. If a sample's absorbance had fallen above 1.98 OD, it would need to be diluted further and re-run rather than reported as an extrapolated value.
Result Interpretation
The R² value indicates how well the curve model fits your standard data. An R² ≥ 0.99 for linear fitting or a visually good sigmoid fit for 4PL means your standard curve is reliable. The EC50 (parameter C in 4PL) represents the concentration at which the assay signal is halfway between its minimum and maximum — a biologically meaningful value that reflects assay sensitivity. OOR flags indicate that the sample absorbance falls outside the standard curve range and the reported concentration should be treated with caution. The "After Dilution" column shows the concentration corrected for any pre-assay dilution and represents the actual concentration in your original sample.
Practical Applications
Use the linear fit option when you are confident your samples fall within the linear portion of the assay — typically the middle section of the sigmoid curve. Use 4PL fit for any assay where standards span more than a 4-fold concentration range or where the curve clearly shows saturation at high concentrations. The 4PL model is the industry standard for cytokine ELISAs, hormone assays, antibody titration, and biomarker quantification.
This calculator is particularly useful for: sandwich ELISAs for cytokines (IL-6, TNF-α, IFN-γ), competitive ELISAs for hormones (cortisol, estradiol), capture ELISAs for antibody quantification, and any assay that requires interpolation from a reference standard curve.
Scientific Notes & Limitations
The 4PL fit uses an iterative gradient-descent algorithm rather than a full Levenberg-Marquardt solver, so unusually noisy or poorly-shaped standard curves may converge to a suboptimal fit — always check the plotted curve and R² value rather than trusting the fit blindly.
The model assumes a single, well-defined analyte-antibody interaction and does not correct for matrix effects, cross-reactivity, or the hook effect (loss of signal at very high analyte concentrations due to antibody saturation). If you see unexpectedly low absorbance at a concentration you expect to be high, dilute the sample further and re-test rather than trusting the interpolated value directly.
Linear fitting is mathematically simple but is only valid across a narrow concentration range near the mid-point of a sigmoid curve; applying it across the full standard range will systematically bias results toward the high and low ends of the curve.
Practical Tips
- Run a fresh standard curve with every plate rather than reusing one from a previous run — day-to-day variation in reagents and incubation conditions shifts absorbance values.
- Include at least one standard near the expected concentration of your samples so results come from interpolation rather than extrapolation.
- Run standards and samples in duplicate or triplicate and enter the mean absorbance, which reduces the impact of pipetting error on the fitted curve.
- For 4PL fitting, choose a standard range that extends well above and below your expected sample concentrations so both asymptotes (A and D) are well defined.
- Keep a record of raw absorbance values before blank subtraction in your lab notebook in case you need to recheck a calculation later.
Common Mistakes to Avoid
- Not subtracting the blank. Always subtract the mean absorbance of your blank (zero standard or buffer-only) wells from all standards and samples before entering values into the calculator. Failure to do so inflates all readings and shifts the curve upward.
- Using too few standards. A minimum of 6–8 standard points is recommended for 4PL fitting. Fewer points may produce a poorly constrained fit, especially at the asymptotes, leading to inaccurate interpolation for samples near the extremes of the curve.
- Ignoring out-of-range samples. Never report concentrations that fall outside your standard curve range. Extrapolation beyond the highest or lowest standard is unreliable. Dilute OOR-high samples and re-run; report OOR-low samples as below the limit of quantification.
- Entering duplicates separately. If you run standards in duplicate or triplicate, enter the mean absorbance for each concentration into the standard table rather than individual replicates, unless the calculator explicitly supports replicate averaging (this one uses mean values per point).
- Mixing up dilution factors. If different samples were diluted to different degrees before plating, calculate each group separately with its own dilution factor rather than using a single factor for all samples.
Frequently Asked Questions
What is the difference between linear and 4PL curve fitting in ELISA?
Linear fitting assumes a straight-line relationship between concentration and absorbance and is only valid within the linear range of the standard curve. Most ELISA standard curves are sigmoid (S-shaped) due to signal saturation at high and low concentrations. The 4-parameter logistic (4PL) model accounts for this sigmoidal shape by fitting four parameters — minimum asymptote (A), Hill slope (B), EC50/inflection point (C), and maximum asymptote (D) — and is the recommended method for sandwich ELISAs that span a wide concentration range. Use linear fit only if you are working in the verified linear portion of your assay.
What R² value is acceptable for an ELISA standard curve?
For a linear standard curve, an R² of 0.99 or higher is typically required to meet assay acceptance criteria in most regulatory and research settings. For 4PL fitting, R² alone is less informative since the model is non-linear; instead, verify that residuals are small and randomly distributed. Many kit manufacturers specify that the standard curve must have a minimum R² of 0.99 for results to be considered valid. If your R² falls below this threshold, check for pipetting errors, expired reagents, or substrates and standards that were not equilibrated to room temperature.
What does "out of range" (OOR) mean in ELISA results?
Out of range (OOR) means the sample absorbance falls outside the absorbance window defined by your lowest and highest standards. Samples with absorbance below the lowest standard may be below the limit of detection (LOD), while samples above the highest standard have exceeded the upper limit of quantification (ULOQ). For OOR-high samples, dilute the sample further and re-run the assay. For OOR-low samples, consider concentrating the sample or using a more sensitive detection method. Never extrapolate ELISA concentrations outside the validated standard curve range.
How should I apply a dilution factor in ELISA calculations?
The dilution factor accounts for any pre-dilution of your sample before it was added to the ELISA plate. For example, if you diluted your serum 1:10 before loading it into the well, enter 10 as the dilution factor. The calculator multiplies the concentration interpolated from the standard curve by the dilution factor to give you the actual concentration in the original undiluted sample. Always use the same dilution factor for all sample replicates in a given run. If different samples were diluted differently, calculate each group separately.
Why should I subtract the blank absorbance before entering ELISA data?
The blank (zero standard or buffer-only well) absorbs light at the detection wavelength due to non-specific binding of the detection antibody and the substrate color before complete stop-solution addition. This background signal inflates all absorbance readings. Subtracting the mean blank absorbance from every standard and sample before entering data into the calculator removes this background and anchors the standard curve to a true zero, which is critical for accurate concentration interpolation — especially for low-abundance samples near the detection limit. Always run duplicate or triplicate blanks and use the mean value for subtraction.