# How do I determine a titre cut-off point?

I am trying to find the type of antibodies in sera of predators by ELISA but at the same time I need to know the titre, however, I need to determine the cut of point to get the titre.

I am trying to find the type of antibodies in sera of predators by ELISA but at the same time I need to know the titre, however, I need to determine the cut of point to get the titre.

- Mathematically, the exact titre is the point of the titration curve where the slope of the tangential line changes its sign, the inflection point (if my English is correct). There are Math application that dervie that for you if you feed your experimental curve. The double derivate of the titration point is THE titre of the sample. It is not subject to errors when different titration curves are not symetrical. This determination allows you to see smaller differences between samples.
- Dear Rauna

In ELISA you will do a dilution range of your antibody sample. At high concentration you will see little difference from one dilution to the next, unless you have a prozone effect. At a certain higher dilution (lower concentration of sample) the plateau of the signal will start to decline significantly. Halfway down to the minimum signal, i.e. at the point of steepest decline of the slope, that is your titre, or cut-off point. If the dilution there was found to be 1 in 2500, then your titre is 2500.

Jan - Run standard along with test or take positive sample which give high od value and use it as standard.
- 1.Carry out a serial dilution of the serum.

2. Conduct the ELISA or any other test using the diluted sera.

3.Plot the OD readings versus the dilution factor. You will observe that the O.D will drop until it reaches the background level.

4.The reciprocal value of the dilution that gives the first background reading is the titre.

Example:

Dilutions 1/5 1/10 1/20 1/40 1/80 1/160 1/320 1/640

Readings 1.2 0.8 0.6 0.4 0.2 0.1 0.05 0.05

.

Titre is reciprocal of 1/320 =320 - buy that antibody detection ELISA kit (Quantitative)...... If kit is not available take known positive sample which can give high od value and do elisa. dilute that sample into 6 dilution and run assay again with same sample...... after that you can take mean average as titre cut-off.
- Hiii, please check the link below, I wish it may help you.

www.mbswonline.com/upload/presentation_10-22-2010-12-9-38.ppt - For finding the antibody titre i think you should serially dilute the serum untill the last value will come consistantly same as another last value. for instance..dilute the serum at 1/200,1/400,1/800,1/1600,1/3200,1/6400,1/12800,1/25600,1/51200...

absorbance of above diluted serum

1.6, 1.4, 1.2, 1.0, 0.8, 0.6, 0.4, ,0.4, 0.4...

The titre cutoff is 1/12800 at 0.4 at OD - Hi.., In ELISA, for calculating titer, you have to take number of dilutions of your sera along with negative control (buffer control). The maximum dilution of sera which shows reactivity to your antigen i.e. your blank o.d. is 0.05 and your maximum dilution of sera gives o.d. 0.1(It shows reactivity of your sera to antigen which is identically higher than negative control), that dilution of sera can be considered as antibody titre for respective antigen.
- Dear Rauna,

Do you know what the target of your antibodies is? If it is a protein or carbohydrate, do you know where it is expressed i.e. is it integral to the membrane, membrane associated, cytosolic or nuclear? - You should choose what you want:

Titration requires a serial dilution. The titer ist equal to the lowest detectable level of your serial dilation of your sample. If you wolud like to have a differentiation between negative and positve samples samples you have to run as much as possible negative samples and positve samples. Calculate the titers of all of them. Now you have to use different cut-off titers and put the results in a four-field table.

test positive test negative

known positive a (right positive) b (false negative)

known negaive c (false negative) d (right negative)

for each cut-off titer. Now you can claculate the sensitivity and the specificity. The last step is the calcluation of the youden-index Y=(Sp-Se)-1 per each cut off titer. The optimal cut-off titer is the onoe wih the highest youden index.

see also: Kießig, S. T. et al. Bestimmung von Schwellenwerten (Cut-off) bei Enzymimmunoassays am Beispiel des FSME-ELISA [Problems of Cut-Off Level Determination in Enzyme Immunoassays: The Case of TBE-ELISA], Klin. Lab., 39, 877 (1993).

original paper: http://www.ncbi.nlm.nih.gov/pubmed/15405679

The much more easier way would be to selct first a dilution which is suitable for all samples. measure the OD and do the same as for the cut-off titeres just here using only an OD as a cut off. (Then you need only one dliution instead of a serial dilution for your samples and you can measure much more samples in a single ELISA) Insteat of the the cut-off OD you can calculate also P/N-ratios (positive/negative).

If you once the cut off, despite of the way you choosed, you can work with the S/C-Ratio (Sample-Cut-off-Ratio) if you have on control smaple adjusted to your cut-off value. - The cut-off paper ist just now online available. Sorry, it's in German.
- Hi,

I think all the advice you've been given is very useful. I agree that you have to make a serial dilution of your samples and also run negative controls. You can use the mean OD of these negative samples plus 3 standard deviations to assign a "cut-off OD". This can then be used to calculate the titre of your samples. The highest dilution of your sample at which the OD is greater than the "cut-off OD" gives your titre e.g. 1:400 dilution = titre of 400.

What is the capture antigen/ antibody in your ELISA? I would warn against using a set dilution of sera for all samples as you may miss important information regarding cross-reactive antibodies or strength of response.

Hope this helps and good luck with the ELISA. - Hi,

There is a common use of cut-off calculation: average of OD of normal serum sample,blank control,buffer control + 3SD. - Thank you very much for your contributions they really help me! Thanks again!

Already a member? Log in

## All Answers (15)