ELISA Concentration Calculator
Calculate analyte concentration from ELISA optical density using standard curve interpolation.
Why 4PL Rather Than a Straight Line
An ELISA standard curve is sigmoid, not linear. It flattens at both ends — at the bottom because signal approaches background, at the top because antibody binding sites saturate.
| Parameter | Meaning |
|---|---|
| A | Lower asymptote — background |
| B | Slope at the inflection |
| C | EC50 — concentration at half-maximal signal |
| D | Upper asymptote — saturation |
Fitting a straight line through the middle points throws away the curve's shape and produces systematic error toward both ends, where most real samples that need dilution decisions actually sit.
Never Extrapolate Beyond the Standards
A sample reading above the top standard cannot be quantified — the curve is flat there, so a large concentration difference produces almost no OD change. The correct action is to dilute and re-assay, not to extend the fit.
Below the lowest standard the problem is different but equally disqualifying: signal is indistinguishable from background. Report as below the limit of quantification rather than assigning a number.
| Reading | Action |
|---|---|
| Above top standard | Dilute and re-assay |
| Below lowest standard | Report as below LOQ |
| Within curve | Quantify from the fit |
| High CV between replicates | Repeat — pipetting or washing problem |
The Hook Effect
At very high analyte concentration in a sandwich ELISA, capture and detection antibodies bind separate analyte molecules rather than bridging the same one. Signal falls, so an extremely concentrated sample can read as low or even normal.
This is the most dangerous ELISA artefact because the result looks plausible. Running every sample at two dilutions detects it: if the diluted sample reads higher than expected from the dilution factor, the undiluted well was hooked.
Edge effects are the other routine problem — outer wells evaporate and equilibrate differently, so critical standards and controls should not be placed there.
Worked Examples
Common Mistakes
The standard curve is sigmoid. Linear fitting introduces systematic error at both ends, where dilution decisions are usually made.
The curve is flat there, so OD barely changes with concentration. Dilute and re-assay instead.
Very high analyte concentrations reduce signal in sandwich assays, so a concentrated sample can read normal. Run two dilutions to detect it.
Outer wells evaporate and equilibrate differently. Keep standards and critical controls away from plate edges.