What a standard curve does
Many assays, such as Bradford, BCA, ELISA and qPCR, give a reading that rises with concentration but not by a fixed, known amount. Measuring samples of known concentration, the standards, at the same time shows how reading relates to concentration in that run, and unknown samples are read off that relationship.
Preparing the standards
- Use 5 to 8 standards spanning the range you expect, including a blank (zero).
- Make them in the same buffer as your samples, so the buffer affects both equally.
- Prepare them by dilution from one stock, and run them in duplicate or triplicate alongside the samples.
Fitting the line
Plot reading against concentration and fit a straight line, reading = slope × concentration + intercept, by least squares. Then for each unknown:
concentration = (reading − intercept) ÷ slope
Multiply by the dilution factor if the sample was diluted before measuring.
A worked example
BSA standards of 0, 250, 500 and 1,000 µg/mL give absorbances of 0.05, 0.37, 0.66 and 1.21. The fitted line is about reading = 0.00115 × concentration + 0.068. A sample diluted 1:5 reads 0.48: (0.48 − 0.068) ÷ 0.00115 = 358 µg/mL in the well, or about 1.8 mg/mL in the original sample.
Judging the curve
- R² above about 0.99 is usual. A lower value points to a pipetting error or an outlier.
- Stay inside the standards. Readings above the highest standard are extrapolated; dilute the sample and measure again.
- Watch for curvature. Many assays flatten out at high concentrations. Drop the top standards that bend away, or use a narrower range.
Calculators
The Standard Curve Calculator fits the line, gives R² and works out every unknown with its dilution. Plan the standards with the Serial Dilution Calculator; for purified proteins, the Protein Concentration Calculator uses A280 instead of a curve.