### Data analysis ELISA

We recommend including a standard curve of 7 dilutions on each ELISA plate, from which a concentration-response relationship is generated.

Prepare the standard curve by plotting the mean OD of each standard dilution (Std 1-7) minus the mean OD of the blank (see formula below) on the y-axis against the corresponding concentration on the x-axis.

Formula: OD = mean OD_{Std 1-7/sample} – mean OD_{blank}

Most laboratories have (plate reader) software that allows various methods of curve fitting. Since ELISA data are essentially sigmoid rather than linear, we recommend using the 4- or 5-parameter logistic fit for quantitative analysis of the samples. Alternatively, a linear regression curve may be acceptable for the linear portion of the curve consisting of at least 3 concentrations.

After selection of the regression model, the analyte concentration in unknown samples can be determined by interpolate the mean OD of the unknown sample minus the mean OD of the blank (see formula above) within the range of standard curve.

Notes:

- The OD value of the sample should fall close to the midpoint between the known concentrations (standard curve). Samples showing an OD below the lowest concentration of the standard curve should be re-analyzed at a lower dilution. Samples showing an OD that exceeds the highest concentration of the standard curve should be re-analyzed at a higher dilution.
- If samples have been diluted, the calculated concentration must be multiplied by the dilution factor.