ELISA Troubleshooting: When Calculated Sample Concentrations Come Back Far Too High

QUICK ANSWER

If your ELISA returns sample concentrations that are orders of magnitude too high, the assay itself is usually fine — the error is almost always in data analysis. The most common culprit is entering the wrong standard concentrations when building the standard curve. In a recent porcine MAP (PigMAP) sandwich ELISA case, a researcher assigned a 5000–78 ng/mL range to standards that should have been 10–0.156 ng/mL. The optical density (OD) readings were valid; only the concentrations tied to them were wrong. Re-assigning the correct values to the same ODs brought results back into a sensible range.

On this page

The problem Snapshot of the data Why it happens Our analysis What we recommended FAQ

The problem

A researcher ran a porcine MAP (major acute phase protein, or PigMAP) sandwich ELISA on serum samples and reached out to our technical support team. After applying dilution factors and converting units, the calculated sample concentrations came back at roughly 150–395 mg/mL. The expected range for these samples was about 0.5–5 mg/mL — meaning the reported values were around 100 times too high.

A follow-up linearity experiment, run across a dilution series, made things more confusing: instead of a clean straight line, the diluted samples produced erratic, non-linear concentrations. The researcher, who described themselves as new to ELISAs, asked us what had gone wrong and how to proceed.

Snapshot of the data

The clue was in the standard curve. The concentrations assigned to the standards were far higher than the kit's validated range. Because every sample concentration is read off that curve, the error cascaded into every result.

Standard concentrations: entered vs. correct

Standard	Entered (ng/mL)	Correct (ng/mL)	Same OD?
Std 1	5000	10	Yes
Std 2	2500	5	Yes
Std 3	1250	2.5	Yes
Std 4	625	1.25	Yes
Std 5	312.5	0.625	Yes
Std 6	156.3	0.3125	Yes
Std 7	78.1	0.156	Yes
Blank	0	0	Yes

Reported vs. recalculated sample concentrations

Metric	As reported	After recalculation	Expected
Sample range	~150–395 mg/mL	~0.24–0.80 mg/mL	~0.5–5 mg/mL
OD values	Valid	Unchanged	—
Curve range used	5000–78.1 ng/mL	10–0.156 ng/mL	10–0.156 ng/mL

Values are illustrative and have been simplified to demonstrate the effect; they do not represent a specific customer result.

Why it happens

A standard curve maps optical density to concentration. Your plate reader measures OD; the concentrations are whatever you tell the software the standards contain. If those concentration labels are wrong, every value calculated from the curve is wrong by the same factor — even when the assay ran perfectly.

In this case the entered concentrations were roughly 500 times higher than the true values. That single mismatch, multiplied by a large dilution factor and a unit conversion, is exactly what pushed the final numbers from a plausible sub-milligram range into the hundreds of milligrams per millilitre. It looked like an assay failure, but it was a data-entry problem.

Our analysis

Cross-checking the standards against the product datasheet showed the validated range for this kit is 10–0.156 ng/mL, not the 5000–78.1 ng/mL that had been entered. We confirmed the kit lot was correct, which ruled out a mislabelled or expired kit and pointed firmly at the analysis step.

Keeping the researcher's OD readings and simply re-assigning the correct standard concentrations, we recalculated the curve and the samples. The corrected concentrations landed around 0.24–0.80 mg/mL — comfortably inside the expected range. The assay had, in fact, worked, and the chosen 1:100,000 sample dilution was appropriate.

The remaining non-linearity in the dilution series had a second, separate cause. Where samples were less diluted, results drifted off the line; where they were diluted further, linearity improved. That pattern is the signature of a matrix effect — sample components interfering with antibody binding when the sample is too concentrated.

ROOT CAUSE

Incorrect standard concentrations were entered when generating the standard curve during data analysis. The OD data and the assay itself were sound. A secondary matrix effect caused non-linearity in the least-diluted samples.

What we recommended

Re-enter the correct standard concentrations (10–0.156 ng/mL) against the existing OD values, then recalculate — no need to re-run the plate.

Always confirm the standard range from the datasheet, kit manual, and box label before building the curve.

For the linearity series, dilute samples further to reduce the matrix effect and restore a straight line.

Treat OD readings below ~0.1 as unreliable for quantification, and report negatives as "not detected" rather than forcing a value.

Confirm the assay format — this is a sandwich ELISA, where higher OD means higher concentration (the opposite of a competitive ELISA).

KEY TAKEAWAY

When ELISA results look impossibly high or low, check the standard curve before you blame the kit. A wrong concentration entered against a correct OD reading will skew every sample by the same factor. For more calculation and curve-fitting guidance, visit our ELISA Support Hub →

Frequently asked questions

My sample concentrations are 100× too high. Do I need to re-run the assay?

Usually not. If the OD readings are valid and the standards were physically prepared correctly, you only need to re-enter the correct standard concentrations in your analysis software and recalculate. The plate data can stay as it is.

Where do I find the correct standard curve range?

It is listed on the product web page, in the kit manual, and on the label on the kit box. Always cross-check all three before assigning concentrations, as the range varies between targets and kits.

What is a matrix effect and how do I fix it?

A matrix effect occurs when components in a concentrated sample interfere with antibody–analyte binding, producing non-linear or inaccurate results. The usual fix is to dilute the sample further until results become linear across the dilution series.

Can I trust very low OD readings?

As a general guide, OD values below about 0.1 sit close to background and should not be used for quantification. Report those samples as "not detected" rather than assigning an unreliable concentration.

Detection range	0.156–10 ng/mL
Sensitivity	0.094 ng/mL
Sample types	Serum, plasma, cell culture supernatant & more

ELISA Troubleshooting: When Calculated Sample Concentrations Come Back Far Too High

The problem

Snapshot of the data

Why it happens

Our analysis

What we recommended

Frequently asked questions

Recent Posts

ELISA Troubleshooting: When Calculated Sample Concentrations Come Back Far Too High

ELISA Troubleshooting: When Standards Read Below the Blank

Introducing BioSymposia: The Gathering Place for Science