Hello,
I am trying to do the Competitive cAMP Complete ELISA Kit from ABCAM.
we are following the instruction of the manual but is not very clear in how to make the calculation.
We would like to know how to do the calculation for standard curve with the Four Parameter Logistic Regression that the manual suggest and how to extrapolate the curve for the samples.
We use the prism 7.0 software to get the results of the curve but we are not sure about the results.
one of our doubt is If we need to transform our values to -log or not.
Does anyone have experience with this kit? or have suggestion to make calculation with competitive ELISA?
Thank you in advance
Pablo M
most of the ELISA kit best to fit in. 4-log parameter axis, and mostly could be done with the program installed. If the program is not installed with the suitable system, a log-log axis is preferred to get a linear standard curved, but data may slightly inadequate. Log-log axis could be done using microsoft excel. Log both OD and the concentration to plot out a linear standard equation(y=ax+b) , next is to Log your sample OD(y) and calculate the (x) with the equation obtained, next using the equation (=10^x) to get the data needed.
most of the ELISA kit best to fit in. 4-log parameter axis, and mostly could be done with the program installed. If the program is not installed with the suitable system, a log-log axis is preferred to get a linear standard curved, but data may slightly inadequate. Log-log axis could be done using microsoft excel. Log both OD and the concentration to plot out a linear standard equation(y=ax+b) , next is to Log your sample OD(y) and calculate the (x) with the equation obtained, next using the equation (=10^x) to get the data needed.
Hi Pablo,
Yu's method is a possible solution to read the concentration from a standard curve linearized this way. Nevertheless, I suggest you to use your Prism software to fit the 4-parameter logistic directly, instead of the linear approximation on a log-log scale. To produce a calculation-stable fit, Prism needs the log of concentration as independent variable (X) instead of the original concentrations. Do not change the measured OD values (Y variable). Prism let you read the unknown concentration, it is a built in functionality. The advantage of using Prism (or any other similar sotfware with non-linear regression) is that you see your logistic curve, instead of just the transformed - linearized log-log values. Only the close-to-linear (middle) part of the standard curve might be used for determination of the unknown concentrations. It is rather straightforward this way, you can avoid the use of the lower and upper curving parts of the concentration-response curve easily.
Cheers, Karoly
Thank you Karoly. However I transformed the values to log and I got weird results, maybe I am not using PRISM correctly for this fit, I followed the suggestion from the website. The linear transformation got me something a bit more reliable and I compared those values with an online website to calculate ELISA values.
Pablo, maybe a bug somewhere in your fitting procedure, hard to figure it out without looking at the data. Nevertheless, you can try other software "good" in nonlinear regression, e.g. Origin (https://www.originlab.com/) or KaleidaGraph (http://www.synergy.com/). I'm using both. These apps do not require any transformation of the data, just introduce the concentration-OD pairs as you obtained them. Both have trial versions. This way you can check if there is a bug in your procedure with Prism...
- Badges
- Science method
- Similar topics
- Medicine
- Immunology
- Immunology Techniques
- Immunoassay
- ELISA