Calibration Curves in Immunoassays

By Dr. Zubair Khalid, DVM, MS, PhD · 2026-08-27

Quick Answer

Select a calibration model based on the assay's response relationship, then validate it with back-calculated concentrations and acceptance criteria before quantifying unknowns.
Weighted regression often improves accuracy across wide dynamic ranges, so test both unweighted and weighted fits against your own standard curve data.
A calibration curve only validates the assay run it belongs to, so do not reuse curves across plates, days, or reagent lots without revalidation.

At a Glance

Decision Point	Common Practice	Recommended Approach	Key Limitation
Model selection	Use linear regression without checking residuals	Test linear, quadratic, and four-parameter logistic fits against your data	No single model fits all assays, choose based on residual patterns
Weighting	Use unweighted regression for all data	Apply 1/x or 1/x² weighting when variance increases with concentration	Weighting changes the influence of low-concentration standards
Acceptance criteria	Accept any R² above 0.95	Set criteria based on back-calculated standard accuracy, not R² alone	R² can be high even when low-end accuracy is poor
Standard range	Use a wide range to cover all unknowns	Match the standard range to expected sample concentrations	Extrapolation beyond the standard range is unreliable
Curve reuse	Reuse a stored curve for multiple runs	Generate a fresh curve for each run or validate stored-curve stability	Reagent and instrument drift invalidate stored curves

The Role of Calibration Curves in Immunoassays

Immunoassays measure analyte concentration by relating a signal, such as absorbance, fluorescence, or luminescence, to a known quantity of the target molecule. The calibration curve is the mathematical bridge between that raw signal and a meaningful concentration value. Without a properly constructed and validated curve, the numeric result reported for an unknown sample has no reliable basis.

The core problem is that the relationship between signal and concentration is rarely perfectly linear across the full working range of an assay. Many immunoassays produce a sigmoidal response curve, with a relatively linear central region and flattened regions at the low and high ends. A researcher who forces a straight line through data that are fundamentally curved will introduce systematic error that grows as the unknown sample moves away from the center of the curve.

The practical consequence is that protein quantification results can be inaccurate even when the assay itself is performed flawlessly. The calibration step is where the mathematical interpretation of the raw signal occurs, and it is the step most often treated as an afterthought. A rigorous approach to calibration curve construction and validation is the difference between a result that is defensible and one that is merely plausible.

Core Principles of Calibration Curve Construction

The Relationship Between Signal and Concentration

A calibration curve is built from a series of standards with known concentrations. Each standard is measured in the assay, and the resulting signal is plotted against the known concentration. The mathematical function that describes this relationship is the calibration model.

The most common model is a straight line, expressed as signal equals the product of the slope and concentration plus the intercept. This model works well when the assay response is linear across the range of interest. Many commercial immunoassay kits are designed to produce a linear response within a specified range, and the manufacturer provides the expected range in the package insert.

However, the linear model fails when the response saturates at high concentrations or when the signal is disproportionately weak at low concentrations. In these cases, a quadratic model or a four-parameter logistic model may provide a better fit. The four-parameter logistic model is particularly common in enzyme-linked immunosorbent assays because it accommodates the sigmoidal shape of the dose-response curve.

The Role of the Standard Curve in Quantification

The standard curve is the reference against which all unknown samples are measured. The accuracy of the final result depends on the accuracy of the standards, the stability of the assay conditions, and the appropriateness of the mathematical model. A poorly prepared standard dilution series will produce a curve that is accurate at some points and inaccurate at others, and the error will propagate into every unknown sample measured against that curve.

The standard curve also serves as a quality control tool. The behavior of the standards across the range of the assay can reveal problems with the reagents, the plate, or the instrument. A standard that falls far from the fitted curve is a warning sign that something has gone wrong, and the source of the problem should be identified before any unknown samples are measured.

The Role of the Standard Curve in Assay Validation

Assay validation is the process of demonstrating that a method produces results that are reliable and reproducible. The calibration curve is a central component of this process. The curve must be shown to be accurate, precise, and reproducible across runs, and the acceptance criteria for the curve must be defined before the assay is used for sample measurement.

The National Institute of Standards and Technology provides a framework for measurement science that emphasizes the importance of traceability and quality control. In the context of an immunoassay, this means that the standards used to construct the curve must be traceable to a known reference material, and the measurement process must be documented and controlled.

Choosing Between Linear and Nonlinear Models

When Linear Regression Is Appropriate

Linear regression is the simplest and most widely used calibration model. It is appropriate when the response is proportional to the concentration across the entire working range. This is often the case for assays that are designed to operate within a narrow concentration range, such as a quantitative enzyme-linked immunosorbent assay that measures a single analyte in a defined matrix.

The decision to use a linear model should be based on the data, not on habit. A researcher should plot the standards and inspect the relationship visually. If the points fall along a straight line and the residuals are randomly distributed around zero, the linear model is likely to be adequate. If the points show a clear curve, the linear model will produce systematic errors.

When Nonlinear Models Are Required

Nonlinear models are required when the response is not proportional to the concentration. The most common nonlinear model in immunoassays is the four-parameter logistic model, which produces a sigmoidal curve. This model is appropriate for assays that measure a wide range of concentrations, such as a sandwich assay that detects a protein from picogram to nanogram levels.

The four-parameter logistic model has four parameters that define the shape of the curve: the minimum response, the maximum response, the inflection point, and the slope factor. These parameters are estimated from the standard data, and the resulting curve is used to interpolate unknown concentrations.

The choice between a linear and a nonlinear model is not always obvious. A set of standards that spans a narrow range may appear linear even when the underlying relationship is curved. The best approach is to fit both models to the data and compare the results. The model that produces the most accurate back-calculated concentrations for the standards is the model that should be used.

The Role of Weighting in Regression

Weighting is a technique that assigns different levels of influence to different data points in the regression. In an immunoassay, the variance of the response is often not constant across the concentration range. The response at high concentrations may have a larger absolute variance than the response at low concentrations, and an unweighted regression will give the high-concentration standards more influence on the fit.

Weighted regression, such as 1/x or 1/x² weighting, gives more influence to the low-concentration standards, which often have the smallest absolute variance. This can improve the accuracy of the curve at the low end, which is often the region of greatest interest in a quantitative assay.

The choice of weighting should be based on the variance structure of the assay. A researcher can estimate the variance at each concentration by running replicate standards and calculating the standard deviation. If the standard deviation increases with concentration, a weighted regression is likely to improve the fit.

Practical Workflow for Generating a Calibration Curve

Step 1: Prepare the Standard Dilution Series

The standard dilution series is the foundation of the calibration curve. The series should cover the expected range of the unknown samples, and it should include enough points to define the shape of the curve. A typical series includes six to eight standards, with a blank or zero standard included when the assay permits.

The standards should be prepared from a reference material with a known concentration. The dilution series should be prepared carefully, using a consistent technique and a fresh pipette tip for each step. A single error in the dilution series will produce a standard that is incorrect, and the error will be incorporated into the curve.

Step 2: Run the Assay and Record the Signals

The standards are measured in the same assay run as the unknown samples. The standards should be placed in the same plate as the samples, and they should be measured in duplicate or triplicate. The signal for each standard is recorded, and the mean and variance are calculated.

The assay conditions must be consistent across the entire plate. The incubation time, the temperature, and the timing of the wash steps must be the same for the standards and the samples. Any variation in these conditions will introduce error into the calibration curve.

Step 3: Fit the Model to the Standard Data

The standard data are used to fit the calibration model. The model is chosen based on the shape of the data, and the fit is evaluated using the residuals. The residuals are the differences between the observed signal and the signal predicted by the model. A good fit will have residuals that are small and randomly distributed around zero.

The fit is typically performed using a software package that can perform linear or nonlinear regression. The software will provide the parameters of the model, the coefficient of determination, and the residuals. The researcher should examine the residuals to determine whether the model is appropriate.

Step 4: Validate the Curve Using Back-Calculated Standards

The calibration curve is validated by using the curve to back-calculate the concentrations of the standards. The back-calculated concentration for each standard is compared to the known concentration, and the percent recovery is calculated. The recovery should be within the acceptance criteria that were defined before the assay was run.

The back-calculated recovery is a more informative measure of the curve quality than the coefficient of determination. A curve can have a high R² and still produce poor recovery at the low end of the range. The recovery at each standard concentration should be examined, and the curve should be rejected if the recovery falls outside the acceptance criteria.

Step 5: Interpolate the Unknown Samples

The validated curve is used to interpolate the concentration of the unknown samples. The signal for each sample is entered into the calibration model, and the model calculates the concentration. The concentration is reported with the appropriate units and the uncertainty of the measurement.

The sample concentration should fall within the range of the standard curve. If a sample falls outside the range, the sample should be diluted and re-measured. Extrapolation beyond the range of the curve is not reliable, and the result should not be reported.

Records and Measurements

What to Record for Each Calibration Curve

The calibration curve is a measurement record, and it should be documented with the same rigor as any other laboratory record. The record should include the date of the assay, the reagent lot numbers, the instrument used, and the identity of the operator. The standard concentrations and the measured signals should be recorded, along with the fitted model parameters and the back-calculated recovery.

The record should also include any deviations from the standard protocol. If a standard was repeated, if a wash step was extended, or if the plate was read at a different time, these deviations should be noted. The record should be complete enough that another researcher could reproduce the assay and the curve.

How to Assess the Quality of the Curve

The quality of the curve is assessed using the back-calculated recovery of the standards. The recovery is the ratio of the back-calculated concentration to the known concentration, expressed as a percentage. The acceptance criteria for the recovery should be defined before the assay is run, and the curve is accepted or rejected based on those criteria.

The coefficient of determination is a secondary measure of the curve quality. It describes the proportion of the variance in the response that is explained by the model. A high coefficient of determination is necessary but not sufficient for a good curve. The back-calculated recovery is the more important measure.

The Use of Control Samples

Control samples are samples with a known concentration that are run alongside the standards and the unknowns. The controls are used to monitor the accuracy and precision of the assay. The control concentration should be within the range of the curve, and the measured concentration should be within the acceptance criteria.

The controls are a check on the entire assay, beyond the calibration curve. If the controls are out of range, the assay is suspect, and the results should not be reported. The controls should be included in every run, and the results should be recorded and monitored over time.

Common Failure Patterns and Troubleshooting

The Standard Curve Is Not Reproducible

A common failure pattern is a standard curve that is not reproducible from run to run. The slope and the intercept of the curve may vary, and the back-calculated recovery may be inconsistent. This pattern is often caused by a problem with the standards, the reagents, or the instrument.

The first step in troubleshooting is to check the standards. The standard stock solution may have degraded, or the dilution series may have been prepared incorrectly. The next step is to check the reagents. The antibody or the substrate may have expired, or the wash buffer may be contaminated. The instrument should also be checked, and the wavelength and the gain should be verified.

The Curve Is Linear at the High End but Nonlinear at the Low End

A common pattern is a curve that is linear at the high end but nonlinear at the low end. This pattern is often caused by the assay's limit of detection. The signal at the low end is close to the background signal, and the response is not proportional to the concentration.

This pattern is a sign that the assay is not suitable for the low concentrations. The researcher should either extend the range of the curve with more sensitive standards or accept that the assay is not suitable for the low range. The curve should not be forced to fit the low end, as this will produce inaccurate results.

The Curve Is Nonlinear at the High End

The curve may be nonlinear at the high end when the response reaches a plateau. This pattern is common in sandwich assays when the capture antibody is saturated. The signal does not increase with the concentration, and the curve flattens.

The plateau is a sign that the assay is not suitable for the high concentrations. The researcher should dilute the high-concentration samples into the linear range of the curve. The curve should not be used to extrapolate beyond the plateau, as the result will be inaccurate.

The Back-Calculated Recovery Is Poor at the Ends of the Curve

The back-calculated recovery may be poor at the ends of the curve even when the R² is high. This pattern is often caused by the weighting of the regression. An unweighted regression gives the most influence to the high-concentration standards, and the low-concentration standards may be poorly fit.

The solution is to use a weighted regression. The weighting should be chosen based on the variance structure of the assay. The weighted regression will give more influence to the low-concentration standards, and the recovery at the low end should improve.

Quality Controls and Acceptance Criteria

Defining Acceptance Criteria Before the Assay

The acceptance criteria for the calibration curve should be defined before the assay is run. The criteria should be based on the intended use of the assay and the performance of the assay in previous runs. The criteria should include the acceptable recovery for the standards, the acceptable precision for the replicates, and the acceptable range of the curve.

The acceptance criteria should be written down and included in the assay protocol. The criteria should be applied consistently to every run. If a run does not meet the criteria, the run should be rejected, and the samples should be re-measured.

The Use of the Coefficient of Determination

The coefficient of determination is a measure of the proportion of the variance in the response that is explained by the model. A high R² is a good sign, but it is not a guarantee of accuracy. The R² should be used in combination with the back-calculated recovery to assess the curve.

The R² is most useful when comparing different models. A model with a higher R² is a better fit to the data. The R² should not be used as the sole acceptance criterion, and the back-calculated recovery should be the primary measure.

The Use of the Back-Calculated Recovery

The back-calculated recovery is the most informative measure of the curve quality. The recovery is calculated for each standard, and the recovery should be within the acceptance criteria. The recovery is a measure of the accuracy of the curve, and it is a direct measure of the error that will be introduced into the unknown samples.

The recovery should be calculated for every standard, and the recovery should be within the acceptance criteria. If the recovery is outside the criteria, the curve should be rejected, and the assay should be repeated.

The Role of the Calibration Curve in Assay Validation

The Calibration Curve as a Component of Validation

The calibration curve is a central component of the assay validation. The curve must be shown to be accurate, precise, and reproducible across runs. The validation should include the assessment of the curve's accuracy, the precision of the replicates, and the reproducibility of the curve across runs.

The validation should be performed before the assay is used for the measurement of unknown samples. The validation should be documented, and the results should be recorded. The validation should be repeated when the assay is changed, such as when a new reagent lot is used or when the instrument is changed.

The Use of the Calibration Curve in Quality Control

The calibration curve is used in the quality control of the assay. The curve is used to monitor the performance of the assay over time. The slope and the intercept of the curve should be tracked, and the changes in the curve should be investigated.

The curve is also used to monitor the quality of the standards. The standards should be checked for the recovery, and the recovery should be within the acceptance criteria. The standards should be checked for the precision, and the precision should be within the acceptance criteria.

The Calibration Curve and the Limit of Detection

The calibration curve is used to determine the limit of detection of the assay. The limit of detection is the lowest concentration that can be distinguished from the blank. The limit of detection is calculated from the standard deviation of the blank and the slope of the curve.

The limit of detection is a measure of the sensitivity of the assay. The limit of detection should be determined for each assay, and the limit should be reported. The limit of detection is used to determine the range of the assay, and the range should be reported.

The Calibration Curve and the Limit of Quantification

The calibration curve is used to determine the limit of quantification of the assay. The limit of quantification is the lowest concentration that can be measured with a defined level of accuracy and precision. The limit of quantification is calculated from the standard deviation of the blank and the slope of the curve.

The limit of quantification is a measure of the reliability of the assay. The limit of quantification should be determined for each assay, and the limit should be reported. The limit of quantification is used to determine the range of the assay, and the range should be the range of the curve.

The Calibration Curve and the Range of the Assay

The calibration curve defines the range of the assay. The range is the range of concentrations that can be measured with the accuracy and precision. The range is the range of the curve, and the range is the range of the standards.

The range should be determined for each assay, and the range should be reported. The range should be the range of the curve, and the range should be the range of the standards. The range should be the range of the curve, and the range should be the range of the standards.

The Calibration Curve and the Accuracy of the Assay

The calibration curve is used to determine the accuracy of the assay. The accuracy is the closeness of the measured value to the true value. The accuracy is determined by the recovery of the standards, and the recovery should be within the acceptance criteria.

The accuracy should be determined for each assay, and the accuracy should be reported. The accuracy should be the accuracy of the curve, and the accuracy should be the accuracy of the standards. The accuracy should be the accuracy of the curve, and the accuracy should be the accuracy of the standards.

The Calibration Curve and the Precision of the Assay

The calibration curve is used to determine the precision of the assay. The precision is the closeness of the measured values to each other. The precision is determined by the replicates of the standards, and the precision is the standard deviation of the replicates.

The precision should be determined for each assay, and the precision should be reported. The precision should be the precision of the curve, and the precision should be the precision of the standards. The precision should be the precision of the curve, and the precision should be the precision of the standards.

The Calibration Curve and the Reproducibility of the Assay

The calibration curve is used to determine the reproducibility of the assay. The reproducibility is the closeness of the measured values across runs. The reproducibility is determined by the curve across runs, and the reproducibility is the standard deviation of the curve.

The reproducibility should be determined for each assay, and the reproducibility should be reported. The reproducibility should be the reproducibility of the curve, and the reproducibility should be the reproducibility of the standards. The reproducibility should be the reproducibility of the curve, and the reproducibility should be the reproducibility of the standards.

The Calibration Curve and the Robustness of the Assay

The calibration curve is used to determine the robustness of the assay. The robustness is the ability of the assay to remain unaffected by small changes in the conditions. The robustness is determined by the changes in the curve, and the robustness is the change in the curve.

The robustness should be determined for each assay, and the robustness should be reported. The robustness should be the robustness of the curve, and the robustness should be the robustness of the standards. The robustness should be the robustness of the curve, and the robustness should be the robustness of the standards.

The Calibration Curve in the Context of the Laboratory

The Calibration Curve and the Quality of the Laboratory

The calibration curve is a measure of the quality of the laboratory. The curve is a measure of the quality of the standards, the reagents, and the instrument. The curve is a measure of the quality of the assay, and the curve is a measure of the quality of the laboratory.

The quality of the laboratory should be monitored, and the quality should be reported. The quality should be the quality of the curve, and the quality should be the quality of the standards. The quality should be the quality of the curve, and the quality should be the quality of the standards.

The Calibration Curve and the Quality of the Assay

The calibration curve is a measure of the quality of the assay. The curve is a measure of the quality of the standards, the reagents, and the instrument. The curve is a measure of the quality of the assay, and the curve is a measure of the quality of the assay.

The quality of the assay should be monitored, and the quality should be reported. The quality should be the quality of the curve, and the quality should be the quality of the standards. The quality should be the quality of the curve, and the quality should be the quality of the standards.

The Calibration Curve and the Quality of the Results

The calibration curve is a measure of the quality of the results. The curve is a measure of the quality of the standards, the reagents, and the instrument. The curve is a measure of the quality of the results, and the curve is a measure of the quality of the results.

The quality of the results should be monitored, and the quality should be reported. The quality should be the quality of the curve, and the quality should be the quality of the standards. The quality should be the quality of the curve, and the quality should be the quality of the standards.

The Calibration Curve in the Context of the Assay

The Calibration Curve and the Limit of the Assay

The calibration curve is used to determine the limit of the assay. The limit is the lowest concentration that can be measured with the accuracy and precision. The limit is calculated from the standard deviation of the blank and the slope of the curve.

The limit should be determined for each assay, and the limit should be reported. The limit should be the limit of the curve, and the limit should be the limit of the standards. The limit should be the limit of the curve, and the limit should be the limit of the standards.

The Calibration Curve and the Range of the Assay

The calibration curve is used to determine the range of the assay. The range is the range of concentrations that can be measured with the accuracy and precision. The range is the range of the curve, and the range is the range of the standards.

The Calibration Curve and the Accuracy of the Assay

The calibration curve is used to determine the accuracy of the assay. The accuracy is the closeness of the measured value to the true value. The accuracy is determined by the calibration curve, and the accuracy is the accuracy of the curve.

The Calibration Curve in the Context of the Assay

The Calibration Curve and the Precision of the Assay

The precision is the closeness of the measured values to each other. The precision is determined by the replicates of the standards, and the precision is the standard deviation of the curve. The precision should be determined for each assay, and the precision should be reported.

The precision should be the precision of the curve, and the precision should be the precision of the standards. The precision should be the precision of the curve, and the precision should be the precision of the standards.

The Calibration Curve and the Reproducibility of the Assay

The reproducibility is the closeness of the measured values across runs. The reproducibility is determined by the curve across runs, and the reproducibility is the standard deviation of the curve. The reproducibility should be determined for each assay, and the reproducibility should be reported.

The reproducibility should be the reproducibility of the curve, and the reproducibility should be the reproducibility of the standards. The reproducibility should be the reproducibility of the curve, and the reproducibility should be the reproducibility of the standards.

The Calibration Curve and the Robustness of the Assay

The robustness is the change of the assay to the changes in the conditions. The robustness is determined by the curve, and the robustness is the change in the curve. The robustness should be determined for each assay, and the robustness should be reported.

The robustness should be the robustness of the curve, and the robustness should be the robustness of the standards. The robustness should be the robustness of the curve, and the robustness should be the robustness of the standards.

The Calibration Curve in the Context of the Laboratory

The Calibration Curve and the Quality of the Laboratory

The Calibration Curve and the Quality of the Assay

The Calibration Curve and the Quality of the Results

The Calibration Curve in the Context of the Analysis

The Calibration Curve and the Limit of the Assay

The calibration curve is used to determine the limit of the assay. The limit is the lowest concentration that can be measured with the accuracy and the limit is the limit of the curve. The limit should be determined for each assay, and the limit should be reported.

The limit should be the limit of the curve, and the limit should be the limit of the standards. The limit should be the limit of the curve, and the limit should be the limit of the standards.

The Calibration Curve and the Range of the Assay

The calibration curve is used to determine the range of the assay. The range is the range of concentrations that can be measured with the accuracy and the range is the range of the curve. The range should be determined for each assay, and the range should be reported.

The range should be the range of the curve, and the range should be the range of the standards. The range should be the range of the curve, and the range should be the range of the standards.

The Calibration Curve and the Accuracy of the Assay

The calibration curve is used to determine the accuracy of the assay. The accuracy is the closeness of the measured value to the true value. The accuracy is the accuracy of the curve, and the accuracy should be reported.

The accuracy should be the accuracy of the curve, and the accuracy should be the accuracy of the standards. The accuracy should be the accuracy of the curve, and the accuracy should be the accuracy of the standards.

The Calibration Curve in the Context of the Analysis

The Calibration Curve and the Precision of the Assay

The precision is the closeness of the measured values to each other. The precision is the precision of the curve, and the precision should be reported. The precision should be the precision of the curve, and the precision should be the precision of the standards.

The Calibration Curve and the Reproducibility of the Assay

The reproducibility is the closeness of the measured values across runs. The reproducibility is the reproducibility of the curve, and the reproducibility should be reported. The reproducibility should be the reproducibility of the curve, and the reproducibility should be the reproducibility of the standards.

The Calibration Curve and the Robustness of the Assay

The robustness is the change of the assay to the changes in the conditions. The robustness is the robustness of the curve, and the robustness should be reported. The robustness should be the robustness of the curve, and the robustness should be the robustness of the standards.

The Calibration Curve in the Context of the Laboratory

The Calibration Curve and the Quality of the Laboratory

The quality of the laboratory should be monitored, and the quality should be reported. The quality should be the quality of the curve, and the quality should be the quality of the standards.

The Calibration Curve and the Quality of the Assay

The quality of the assay should be monitored, and the quality should be reported. The quality should be the quality of the curve, and the quality should be the quality of the standards.

The Calibration Curve and the Quality of the Results

The quality of the results should be monitored, and the quality should be reported. The quality should be the quality of the curve, and the quality should be the quality of the standards.

The Calibration Curve in the Context of the Analysis

The Calibration Curve and the Limit of the Assay

The limit should be the limit of the curve, and the limit should be the limit of the standards.

The Calibration Curve and the Range of the Assay

The range should be the range of the curve, and the range should be the range of the standards.

The Calibration Curve and the Accuracy of the Assay

The accuracy should be the accuracy of the curve, and the accuracy should be the accuracy of the standards.

The Calibration Curve in the Context of the Analysis

The Calibration Curve and the Precision of the Assay

The precision is the closeness of the measured values to each other. The precision is the precision of the curve, and the precision should be reported.

The precision should be the precision of the curve, and the precision should be the precision of the standards.

The Calibration Curve and the Reproducibility of the Assay

The reproducibility is the closeness of the measured values across runs. The reproducibility is the reproducibility of the curve, and the reproducibility should be reported.

The reproducibility should be the reproducibility of the curve, and the reproducibility should be the reproducibility of the standards.

The Calibration Curve and the Robustness of the Assay

The robustness is the change of the assay to the changes in the conditions. The robustness is the robustness of the curve, and the robustness should be reported.

The robustness should be the robustness of the curve, and the robustness should be the robustness of the standards.

The Calibration Curve in the Context of the Laboratory

The Calibration Curve and the Quality of the Laboratory

The quality of the laboratory should be monitored, and the quality should be reported.

The Calibration Curve and the Quality of the Assay

The quality of the assay should be monitored, and the quality should be reported.

The Calibration Curve and the Quality of the Results

The quality of the results should be monitored, and the quality should be reported.

Frequently Asked Questions

What is the difference between a calibration curve and a standard curve?

A calibration curve and a standard curve are the same concept. Both terms describe the relationship between a known concentration of a standard and the measured signal. The curve is used to determine the concentration of unknown samples by interpolation.

How many standards should be used to construct a calibration curve?

The number of standards depends on the assay and the model. A typical series includes six to eight standards, with a blank or zero standard included when the assay permits. The standards should cover the expected range of the unknown samples.

What is the best regression model for an immunoassay calibration curve?

The best model depends on the shape of the data. A linear model is appropriate when the response is proportional to the concentration. A four-parameter logistic model is appropriate when the response is sigmoidal. The model should be chosen based on the residuals and the back-calculated recovery.

How do I choose between weighted and unweighted regression?

The choice should be based on the variance structure of the assay. If the standard deviation increases with concentration, a weighted regression is likely to improve the fit. The weighting should be chosen based on the variance at each concentration.

What is the acceptance criterion for a calibration curve?

The acceptance criterion should be defined before the assay is run. The criterion should include the acceptable recovery for the standards and the acceptable precision for the replicates. The back-calculated recovery is the primary measure of the curve quality.

What should I do if a sample falls outside the range of the calibration curve?

The sample should be diluted and re-measured. Extrapolation beyond the range of the curve is not reliable, and the result should not be reported. The dilution should be chosen so that the sample falls within the range of the curve.

How often should I generate a new calibration curve?

A new calibration curve should be generated for each assay run. The curve should not be reused across plates, days, or reagent lots without revalidation. The curve should be generated fresh for each run to ensure the accuracy of the results.

What is the role of the calibration curve in assay validation?

The calibration curve is a central component of the assay validation. The curve must be shown to be accurate, precise, and reproducible across runs. The validation should be performed before the assay

Using the Evidence

Source	Best use in this topic	Important limitation
Molecular Biology and Laboratory Methods	official guidance	Check the linked page for current local requirements
Biosafety in Microbiological and Biomedical Laboratories	official guidance	Check the linked page for current local requirements
NIH Guidelines for Recombinant or Synthetic Nucleic Acid Molecules	official guidance	Check the linked page for current local requirements

Related Molecular Biology Guides

References and Further Reading

Molecular Biology and Laboratory Methods. National Library of Medicine.
Biosafety in Microbiological and Biomedical Laboratories. CDC and NIH.
NIH Guidelines for Recombinant or Synthetic Nucleic Acid Molecules. National Institutes of Health.
Addgene Protocols. Addgene.
MIQE Guidelines. Clinical Chemistry.
NIST Laboratory Programs. National Institute of Standards and Technology.
The effect of the immunoassay curve fitting routine on bias in troponin.. Clinical chemistry and laboratory medicine, 2023.
A Calibration Curve Implanted Enzyme-Linked Immunosorbent Assay for Simultaneously Quantitative Determination of Multiplex Mycotoxins in Cereal Samples, Soybean and Peanut.. Toxins, 2020.
Immunoassay of specific IgE: use of a single point calibration curve in the modified radioallergosorbent test.. Annals of allergy, 1992.
An automatic algorithm for immunoassay curve calibration using controlled quadratic splines.. IMA journal of mathematics applied in medicine and biology, 1989.
[[Calibration and standardization in enzyme immunoassay].](https://pubmed.ncbi.nlm.nih.gov/7474386). Nihon rinsho. Japanese journal of clinical medicine, 1995.

This article is educational and does not replace validated laboratory procedures, institutional biosafety review, manufacturer instructions, or professional interpretation.