1. Input ELISA data of standard into software

After running up the ELISA assay, you can separate your raw data into three parts. It is recommended to run standards and samples in duplicate or triplicate. As a best practice, you'd better control the CV of replicates less than 8%.

• Part 1: absorbance of standards with known concentration
• Part 2: absorbance of blank well/background
• Part 3: absorbance of samples with unknown concentration

Standards are serially diluted with Sample Diluent. Samples may also need to be diluted. Since sample diluent has absorbance even when there is no protein present at the detection wavelength, we run blank well with sample diluent in the assay to get the background OD. It is common practice to subtract the absorbance of the blank well from standards and all sample absorbance.

Take Human TNF-α ELISA kit (CSB-E04740h) as an example. Absorbance of both standards and samples are corrected by subtracting OD of blank well.

Here we use "Curve Expert 1.4" as an example to show you how to operate the software and process data analyzing.

1.1 Start "Curve Expert 1.4". You can see an interface below.

1.2 Input OD values (X-axis) and corresponding concentration (Y-axis) of standards. You can copy the data from excel sheet directly to the worksheet in the software. Data plot appears in the lower right corner.

2. Choose the best fitting curve

2.1 Click [ Run ] button and let software examine your data to choose the best possible curve fit. Select the model families to include in calculation. Usually, we click [ All On ] to include all model families. If the polynomial family is included for consideration, you must specify the maximum degree of the polynomial that software will consider in the "Polynomial Constraint" area. We recommend setting the maximum degree of polynomial as “4”. Of course, if the polynomial family is not included, the polynomial constraint will simply be ignored. Then press [ OK ].

2.2 The calculation will then proceed, and each curve fit is ranked according to its standard error and correlation coefficient, and the best fit is then displayed in the graphing window.

Though the software will recommend the best fitting curve, you can still select curve manually by double clicking different regression. Select the curve with “r” value closest to “1”.

Selecting an appropriate curve-fitting regression is important since concentration of samples will be calculated according to the standard curve. The following figures illustrate how different ELISA standard curves affect the accuracy of concentration calculations. The closer “r” value is to “1”, the stronger correlation there is between OD and concentration.

This is an example to show you how different standard curves affect result. If a test sample produces an absorbance of 1.4, the corresponding two concentrations calculated by linear curve (r= 0.92255796) and nonlinear curve (r= 0.99993479) have a huge difference (199.955 VS 114.898).

In this example we finally select “Rational Function” since its “r” value is closest to “1” compared with other standard curves. By pressing the right mouse button or by pressing Ctrl-M in any graphing window, you can pick from the graphing menu. By clicking [ Copy ] button you can paste standard graph to excel sheet or word document.

3. Calculate the concentration of target protein by interpolation

How to calculate concentration? You have two ways to calculate the concentration of target protein.

• Analyze ELISA data in the software.

The software allows easy location of x/y points, differentiation, and integration of a curve fit that has been performed by choosing Analyze from the graphing menu or pressing Ctrl-L. In this example we would like to find the concentration of target protein (Y value) by evaluating the curve fit at a given OD (X value). Just type in the point at which you want to evaluate the curve fit (the x value) in the "At X=" field and press [ Calculate ] or [ Enter ] on your keyboard. The calculator will then show the corresponding y value.

• Analyze ELISA data in the MS Excel sheet.

Press [ Info ] button to open model information dialog which gives all of the information needed to apply a curve fit and some supplemental information about the performance of the given model. The most important information contained in this panel is the values of the parameters of the model.

Model

The Model section gives the data modeling equation so that the interpretation of the coefficients is clear. The coefficients a, b, c, etc. in the Coefficients section match the corresponding parameters shown in the model.

Coefficients

The Coefficients section gives the values of the relevant parameters for the current model -- these coefficients are always expressed as a, b, c, d, and so on. These coefficients match the model shown in the Model section. If the list of coefficients exceeds the extent of the window that they are shown in, a scrollbar will appear to allow you to scroll through the parameters.

If we click [ Copy ], the list of coefficients is copied to the clipboard. Then paste them in any Excel sheet

Bring X (OD), a, b, c, d value into the formula respectively and then press [ Enter ] on your keyboard to calculate concentration. To get the most accurate results, it is recommended to do a pre-text to determine the appropriate dilution factor to make sure that the concentration of target protein in the samples after dilution falls into the range of standard curve. If the sample has been diluted, the dilution factor should be taken in to account during calculation.

Standard curve is so important to get ELISA results analysis accurately. Therefore, you should run standards in each assay to avoid systematic deviation caused by difference on operator, pipetting, incubations, and temperature of each assay.