We obtained OD of 400 samples from vaccinated fish and control groups.
How can we recognize immunity levels in examined fish?
I believe you need to follow the manufacturer's instruction, otherwise you should compute a cut off or threshold for your, self constructed, ELISA. I'm not sure you'd purchased the ELISA kits or constructed yourself. By the way you would need a quantitative ELISA kits rather than qualitative ones.
Dear Swati,
This study was carried out according to new ELISA design and our setting up in our Laboratory.
How to calculate cut-off point and distinguish level of Antibody production?
Thanks again
Regards
I understand that you are detecting specific antibodies. You can analyze your results in OD or you can determine the antibody titer. The last option is better.
If you work with unique dilution of the samples you can determine the cut-off value with the OD of samples of control group (median +/- 2 or 3 SD).
You need to perform serial dilution of your samples if you want to determinate titers them.
In both cases you need to analyze your datas with a satistical test in order to probe if the differences between groups are significant or not.
The simple way is to use an cut off.
How to determine this:
1. optimize your assay to the the highest P/N-Ratio (OD positive sample / OD negative sample). You can used pooled serum samples. Optimize your antigen concentration at the solid phase, sample dilution, buffer: PBS with 0,3 mol/L NaCl, 1% BSA, 0,05% Tween 20, incubation times should be at 90 min, conjugate concentration (in the same buffer),
2. check a defined set of samples (as much as possible negative and positive). Go for biometry, check the distribution of your values. in most cases the Gaussian bell curve could not be applied, therefor you have to look for the lowest point where both distribution curves intersect. That#s the first hint for an optimal cut off.
3. Now vary this potential cut off value step by step. Calculate for each of this cut off values the specificity and the sensitivity of your assay.
True positive = correctly identified
False positive = incorrectly identified
True negative = correctly rejected
False negative = incorrectly rejected
sensitivity=true positive/(correctly identified + incorrectly identified)
speciticity=True negative/(correctly rejected + incorrectly rejected)
and now the trick:
use the Youden index
Y=(specificity+sensitivity)-1
If you draw this index vs. the cut off values you will get an optimum. That's your point for the best differentiation between normal and vaccinated ...
(Cancer. 1950 Jan;3(1):32-5. Index for rating diagnostic tests. YOUDEN WJ.)
4. This will work until your OD's are stable. To overcome the day-to-day variations you can use a kind of short calculation/adjustment: use a defined negative control at each assay/plate. So you can calculate a sample/negative control-ratio (positive/negative-ratio, P/N-Ratio). Start with point 2. I suggest to prepare before anything starts, to prepare a pool of negative serum sample (and if available also a pool of positive samples), so that you can use them for test optimization as well as for negative and positive control on each test plate for a longer time period.
- Badges
- Science method
- Similar topics
- Medicine
- Immunology
- Immunology Techniques
- Immunoassay
- ELISA