I've performed an ELISA on conditioned media of protein secretion. This assay was done on a serially diluted sample set with dilution factor of 6. However, when accounting for the dilution during analysis of the data after determining the sample concentration from the standard curve, the sample concentrations are not consistent across the range the assay should be able to detect. This raises concern about the kit/my analysis. Any tips/guidance on this issue would be greatly appreciated.
Difficult to answer fully with the limited amount of information provided, but generally: all ELISAs have a range over which they generate linear data, and this range depends on the antigen, the antibody, the method of detection and a host of other minor variables. Really good ELISAs can give useful linear information over two orders of magnitude, while others may only be capable of linearity over a very narrow range (even as low as 2-fold).
It is also important to note that most commercial elisa kits are optimised for specific conditions (generally, conditions under which that elisa behaves at its absolute best). Your own conditions may involve different media, different pH values, different ion concentrations and compositions and all these other things that can alter the behaviour of antibodies, so it is always important to establish a priori the behaviour of your kit under your own specific scenarios. Don't assume that just because the kit SHOULD detect X at Y dilution after Z hours, it actually will.
I typically perform a ten-fold dilution series just to determine the ballpark range (i.e. somewhere between 'no signal' and 'saturated signal'), then employ a second two-fold dilution series over that ballpark range to really pin down the linear range.
Hi Serey,
It is difficult to guess what is the specific problem from the information in your question. I assume that your objective is to do quantification of some analyte by ELISA using standards.
If the kit suggested that you should expect linear values over a range of quantity of standard and if you are able to get it, then the problem could be in the way you are diluting the samples.
Probably you may try using the same diluent for the standard as well as the analyte.
You may also try plotting linear/linear, log/linear or log/log graph between the concentrations and the observed ODs and see if you can resolve the problem of linearity.
If not then you may also try the nonlinear regression such as 4 parameter Log agonist versus response, which may solve the problem of linearity of your standards/analyte.
Nonlinear curve fitting can be performed in GraphPad prizm.
Good luck,
S
- Badges
- Science method
- Similar topics
- Medicine
- Immunology
- Immunology Techniques
- Immunoassay
- ELISA