Best fitting curve for ELISA standard Curve?

Antibody Based Technologies Recombinant Antibody Technology
1

Hi, Is the linear regression the best for the ELISA STD curve or others as the 4-parameter fit are better?

Thanks

2

Daniel nonlinear regression models, particularly the 4-Parameter Logistic (4PL) or 5-Parameter Logistic (5PL) models, are considered far superior for fitting ELISA standard curves.

  • ELISA responses (optical density vs. concentration) follow a sigmoidal (S-shaped) curve, not linear, especially over a wide dynamic range.
  • Linear regression only works over a narrow concentration range where the response is approximately linear—typically the mid-log portion of the curve.
  • Using linear regression risks significant error at high and low concentrations, leading to inaccurate sample interpolation.
2 Likes
3

Daniel

You can also add a fifth parameter to allow for asymmetry in the curve. It’s Used when your ELISA response isn’t symmetric — but 4PL is usually good enough.

4 Likes