ELISA Hormone Assay

[Dr.Stein]

Hello,

I am new in ELISA hormone assay. I am doing a research about measuring testosterone level in mice treated by some substance. Unfortunately our ELISA reader cannot create automatic result, thus I need to determine it semi-manually according to the OD. I am already created the standard curve using Excel and now I have no idea how to determine the testosterone level of my samples. I know that it can be harvested from extrapolation based on the standard curve but is there any automatical generation? Because if we generate it manually with ruler the result will be not valid because of the logarithmic scale and the X-axis interval (it is 0; 0.2, 0.5, 1, 2, 6, 16). For instance if the extrapolation is somewhere between 6 and 16, it would be hard to say the hormone level, right? Though there is some minor grid, but still can't be precise as it should be.

Thank you for your help!

[sdekivit]

expand your calibration curve with higher testosterone concentrations and/or greater concentration steps. And then interpolate.

[Dr.Stein]

I DO KNOW that interpolation but it will provide me not exact concentration number because it will show me a point in a certain range. I already applied this and I am not satisfied, also my supervisor demands an exact number from calculatiion not just a result from a ruler.

For instance, I got a point somewhere between the concentration 6 and 16 ng/mL. How can I get the exact number? And also that the semi-log paper doesn't provide minor scale I need to determine number between those 6 - 16 ng/mL.

I am still confused

I remember that a couple years ago I worked on ELISA in the lab which can calculate automatically in computerized thing soon after we got the OD from the reader. Do you think that I can put my OD data which took from another reader to that program/computer and calculate it there?

I am soo bad at math. I cannot deal well with numbers and equation, but I do not give up yet. I keep trying to solve this calculation. If anyone knows how to do, please teach me. I will be the happiest person in the Earh

[blcr11]

The CDC has software available for ELISA analysis. I think it's even free.

http://www.cdc.gov/ncidod/dbmd/bimb/elisa.htm


Also, a 1986 paper on the methods used to establish the software:

http://www.cdc.gov/ncidod/dbmd/bimb/ELI ... erence.pdf


It's difficult to offer specific advice since there are different ways to fit the standard curve, but once you have, the process is basically as you've both said.


The process of interpolation is not error-free. If you want to reduce the errors of estimation, I think the only two things you can do is to use more points and to increase the number of replicates per point on the standard curve. Increasing the number of replicates per unknown sample can help reduce errors too, but that is only increasing the precision of the unknown measurement, which isn't changing the interpolation errors per se, but is reducing the range of possible values for the analyte level.

If the software you mention can read your OD data, so long as you have the data in a format the program expects, I would think you could analyze your data with that software, too. Obviously, I can't be sure, but it's worth trying.

[fcs]

Hitachi Software MiraiBio Group has a pretty nice elisa analysis software called MasterPlex ReaderFit and there is even a fully functional 14-day trial that you can download for free.

Here is a movie demonstrating a sample elisa analysis with MasterPlex ReaderFit.

Also, here are a few tips for elisa analysis.