Zubair Khalid

Virologist/Molecular Biologist | Veterinarian | Bioinformatician

Conventional & Molecular Virology • Vaccine Development • Computational Biology

Dr. Zubair Khalid is a veterinarian and virologist specializing in conventional and molecular virology, vaccine development, and computational biology. Dedicated to advancing animal health through innovative research and multi-omics approaches.

Dr. Zubair Khalid - Veterinarian, Virologist, and Vaccine Development Researcher specializing in Computational Biology, Multi-omics, Animal Health, and Infectious Disease Research

Blog · Guides · Published 2026-07-12

ELISA Workflow: Designing Controls and Interpreting Plate Results

An enzyme-linked immunosorbent assay (ELISA) is a plate based detection technique used in research and clinical diagnostics. To obtain reliable quantitative data, you must design proper controls, optimize plate layout, and apply correct curve fitting methods. This guide is for anyone performing ELISA experiments who needs to ensure reproducibility and accuracy. For foundational principles, see the NCBI Bookshelf.

The core steps include coating, blocking, washing, detection, and signal readout. However, the difference between a publishable result and a failed experiment often lies in how you handle the pre assay planning. Training resources like EMBL-EBI Training provide modular courses on immunoassay design. This guide focuses on the critical decisions that transform raw absorbance values into defensible concentration estimates.

At a Glance

Element Purpose Key Recommendation
Standards (calibrators) Define the relationship between signal and concentration Use at least 6 points covering the expected dynamic range
Blank wells Measure background signal Include both assay blanks (no sample or antibody) and matrix blanks (sample diluent only)
Replicates Estimate technical variability Use triplicate wells for standards and duplicates for unknown samples
Plate layout Minimize systematic spatial errors Place standards in a column, randomize unknown samples, and consider edge effects
Curve fitting Interpolate unknown concentrations Log transform standards, then fit with a four parameter logistic (4PL) or five parameter logistic (5PL) model
Plate effects Correct for row and column biases Include internal controls in each plate position during validation
Validation Confirm specificity, precision, and accuracy Run spiked recovery and dilution linearity tests using relevant sample matrices

Decision Criteria for Control Design

Every ELISA must include several control types. The choice depends on your sample matrix and assay format. Use the following criteria to decide which controls are essential.

Assay blanks contain all reagents except the analyte. They correct for nonspecific binding of detection antibodies to the plate. For sandwich ELISA, include a blank where the coating antibody is present, but the detection antibody is omitted to assess cross reactivity.

Matrix blanks use the diluent that matches your sample (for example, 1% BSA in PBS for serum samples). They account for background interference from the sample buffer itself. In practice, you should see matrix blank values that are indistinguishable from assay blanks. If they are higher, your diluent may contain contaminating substances or binding proteins. Sources like Galaxy Training Network emphasize the importance of defining a proper blank control for any quantitative bioassay.

Positive controls are samples of known concentration spiked into the same matrix as your unknowns. They confirm that the assay is working correctly. Use at least two levels (low and high) within the standard curve range. For example, a sandwich ELISA for carcinoembryonic antigen related cell adhesion molecule 5 used spiked recombinant protein as a positive control according to Int J Biol Macromol.

Negative controls include samples known to be devoid of the analyte. For patient samples, this could be a well characterized negative pool. Include one negative control per plate to monitor for false positives.

QC samples are aliquots of a pooled sample that you freeze and run with every plate. They allow you to track inter assay variability over time. Maintain a Levey Jennings chart for these controls to detect trends or shifts in performance.

Practical Workflow for Designing Controls and Interpreting Results

1. Plan the Plate Layout

Draw a plate map before pipetting. Dedicate one column (8 wells) for the standard curve. Use duplicate or triplicate wells for each standard dilution. Place blanks in at least four wells, ideally in separate corners of the plate to assess spatial uniformity. Arrange unknown samples in a random order across the remaining wells. Avoid placing all unknowns of the same condition in adjacent columns, as this can confound edge effects.

2. Prepare the Standard Curve

Reconstitute your standard according to the manufacturer or validated protocol. Make serial dilutions in the sample diluent. For a typical sandwich ELISA, you might start at 1000 pg/mL and perform twofold dilutions down to 15.6 pg/mL. Include a zero standard (diluent only) to define the lower bound. Use the same pipette tips and mixing protocol for each dilution to minimize error. When working with complex biological matrices, consider using a standard from the same species to avoid matrix mismatch, as noted in J Med Virol.

3. Include Replicates

Technical replicates capture pipetting and reading variability. Run each standard in triplicate. For unknown samples, duplicates are usually sufficient if the CV between replicates is below 15%. If you expect high variability (for example, from tissue homogenates), increase to triplicates. Record each replicate as a separate well, do not read the same well multiple times and average because that only measures instrument noise, not experimental error.

4. Run the Assay

Follow your validated incubation times and temperatures. Cover the plate during incubation to prevent evaporation, especially at the edges. Wash steps are critical. Use an automatic washer if available, and verify that the wash buffer is completely removed after each cycle. Incomplete washing leads to high backgrounds and poor curve fit.

5. Fit the Standard Curve

After reading absorbance, subtract the mean blank value from all raw signals. Do not subtract blanks from standards if the blanks are in a different matrix. Then transform the standard concentrations (usually log10) and fit a curve. The four parameter logistic (4PL) model is the standard for most ELISA data. Use software like GraphPad Prism, R with the drc package, or commercial ELISA readers that include 4PL fitting. Evaluate the fit by checking that the residuals are random and that the R squared is above 0.99. Poor fits often indicate a bad dilution series or an outlier well.

6. Interpolate Unknown Concentrations

Use the fitted curve to back calculate concentrations from the average absorbance of each unknown sample. If a sample absorbance falls below the lowest standard, report the result as below the detection limit. If it exceeds the highest standard, dilute the sample and re run. Apply the dilution factor after interpolation. Always report the CV among replicates and flag any values with CV above 20%.

7. Validate the Results

Run at least one spiked recovery experiment during plate validation. Add a known amount of analyte to several sample matrix types and measure recovery. Acceptable recovery is between 80% and 120%. Also perform a dilution linearity test by serially diluting a high sample and confirming that the measured concentration changes proportionally. These steps are described in detail for molecular assays in EJNMMI Radiopharm Chem, and the same principles apply to ELISA.

Common Mistakes

Using single wells for standards. A single well per dilution cannot distinguish between an outlier and a real signal. Always use at least duplicates, preferably triplicates, for the standard curve.

Ignoring plate edge effects. Wells at the edges of a 96 well plate often show higher or lower absorbance due to uneven temperature or evaporation. To mitigate this, avoid placing critical unknowns in the outer ring of wells. Use those positions for blanks or buffer only. If your protocol requires full plate occupancy, use a plate seal and incubate in a humidified chamber.

Selecting the wrong curve fit. Many researchers apply a linear fit to log transformed data. While convenient, this assumption is rarely valid over the full dynamic range. Use a 4PL or 5PL model instead. The 5PL model adds an asymmetry parameter, which can improve fit when the curve is not symmetrical. Bioconductor provides packages for robust curve fitting in R.

Forgetting to validate the sample matrix. A standard curve prepared in buffer may not reflect the behavior of the analyte in serum, plasma, or cell lysate. Matrix components can suppress or enhance signal. Always prepare standards in the same matrix as your unknowns when possible. If that is not feasible, perform a spike recovery experiment to document any matrix effect.

Limits and Uncertainty

Every ELISA has quantifiable limits. The limit of detection (LOD) is typically defined as three standard deviations above the mean of blank readings. The limit of quantification (LOQ) is the lowest concentration that can be measured with acceptable precision (CV less than 20%). Report these values for each assay run.

Uncertainty arises from several sources. Pipetting error, temperature gradients during incubation, and reader variability all contribute. A well designed plate layout can minimize systematic error, but random error remains. The best estimate of uncertainty is the CV of replicate wells. For inter assay variability, include a QC sample in every plate and calculate the coefficient of variation across plates. If the inter assay CV exceeds 20%, your assay requires further optimization.

Biological variability is a separate concern. ELISA measures total immunoreactive protein, which may not correlate with functional activity. Use orthogonal methods (such as Western blot or mass spectrometry) to confirm specificity when developing a new assay. For example, Methods Enzymol describes lectin based assays that combine ELISA format with specific glycan recognition, validating results with complementary techniques.

Frequently Asked Questions

Q: How many standard points do I need for a reliable curve?
A minimum of six non zero points is standard. Fewer points may miss the linear region of the sigmoidal curve, leading to poor interpolation near the asymptotes.

Q: Should I subtract the blank from every well?
Yes, but only if the blank matrix matches the standard and sample matrix. If blanks are in a different diluent, subtract a matrix blank instead of the assay blank.

Q: My standard curve has a high background. What went wrong?
High background often results from incomplete washing, excessive antibody concentration, or insufficient blocking. Re evaluate your wash steps and consider increasing the number of wash cycles.

Q: Can I reuse a standard curve across multiple plates?
No. Each plate should have its own standard curve because plate to plate variability in incubation time or temperature will shift the curve. Always run fresh standards with each plate.

References and Further Reading

Related Articles