ELISA & BCA Standard Curve Calculator
Linear regression standard curves for BCA, Bradford, ELISA and qPCR with R², blank subtraction and batch unknown interpolation.
How it works
Standards are fitted by ordinary least squares after blank subtraction, returning slope, intercept and R². Unknowns are interpolated from the fitted line and flagged when they fall outside the standard range, where extrapolation is unreliable. For qPCR standards, amplification efficiency is derived from the slope as E = 10^(-1/slope) − 1.
How linear and 4PL fits work in this tool
For Bradford, BCA, Lowry, and linear ELISA windows, the calculator fits y = slope · x + intercept by ordinary least squares on blank-subtracted readings, then solves x = (y - intercept) / slope for each unknown and flags any sample outside the standard range.
For binding assays and ELISAs that bend toward lower and upper plateaus, it fits the four-parameter logistic curve y = d + (a - d) / (1 + (x / c)^b) across a coarse 4D grid followed by 40 coordinate-refinement passes.
- Both asymptotes a (bottom) and d (top) are searched beyond the observed min and max standards instead of pinned to the highest standard reading. Pinning d to the top standard bends the curve downward and causes division by zero when an unknown reaches that OD.
- Inversion uses x = c · ((a - d) / (y - d) - 1)^(1 / b). Unknowns at or beyond either fitted asymptote return an explicit above_range or below_range notice instead of a silent 0 or extrapolated number.
Worked examples, computed by the tool’s own functions
- Linear regression on (x, y) pairs (1, 2), (2, 3), (3, 5), (5, 6): slope = 1.0286 (36/35), intercept = 1.1714, R² = 0.9257.
- 4PL calibration from standards at 0, 2, 10, 31.25, 62.5, 100, 250, 500, and 1000 generated with true (a = 0.05, d = 2.5, c = 100, b = 1.2): the fitter recovers a = 0.050, d = 2.499, c = 99.87, b = 1.201, and R² = 1.0000.
- Midpoint unknown OD = 1.275 inverts to concentration 99.98; saturated unknown OD = 2.6 returns above_range rather than an unmeasured extrapolate.
Frequently asked questions
How do I calculate protein concentration from a BCA standard curve?
Subtract the blank absorbance from every reading, plot the known standard concentrations against their corrected absorbance, and fit a straight line. Rearranging that line gives concentration = (absorbance − intercept) ÷ slope. Only interpolate within the range your standards covered — samples above the top standard must be diluted and re-read rather than extrapolated.
What R² is acceptable for a standard curve?
For most colorimetric protein assays, R² above 0.98 is expected and above 0.99 is routine. A lower value usually points to pipetting error, an expired reagent, or readings outside the assay’s linear range rather than to genuine curvature. R² alone is not sufficient — always check that residuals are scattered randomly rather than showing a systematic bend.