Abour PCR sensitivity
djones at SCRI.SARI.AC.UK
Wed Apr 24 06:55:03 EST 1996
Juan Pablo Martinez-Soriano replied to Kevin's earlier message
> > The work was carried out using purified
> >virus and viral RNA rather than tuber material, however that will be the
> >next stage. The paper reports a 10-fold increase in sensitivity over
> >ELISA. We have since increased that to 100-fold and believe that there is
> >further room for improvement.
> In recent experiments made at our lab, we have improved RT-PCR sensitivity
> up to 1000-fold when compared with DAS-ELISA trying to detect Citrus
> Tristeza Virus from FIELD samples. We have been using 2 sets of primers
> (four oligos) that amplify two different regions of the viral genome but
> yield same size fragments.
> We have seen that sensitivity goes even way up if we combine ELISA and
> RT-PCR. That is, RT-PCR made after ELISA.
> >For many purposes it just doesn't make sense at
> >the moment to use nucleic-acid based techniques when cheaper ones do
> >the job just as well.
> >However, post harvest tuber testing in potatoes may be an exception,
> Well, you said it well. It depends on what is the bottom end of the
> detection. In Mexico we are doing a massive survey where thousands of trees
> need to be tested for CTV. It would be crazy to even thing to try RT-PCR on
> each of them. We pool 5 to 10 samples (each coming from one tree) and do
> ELISA on them, which was fine for some time. Suddenly, number of samples
> and time became critical and we needed more fast and sensitive tools. We
> know now that we can pool 100 (or more) samples and detect the presence of
> one infected tree (even if 99 are healthy) with our RT-PCR. You can easily
> see that cost is divided by 100 (or more) making the PCR test (per tree)
> way cheaper than ELISA. Do not ask me why we do not do hybridisations.
> See, potaoes are not exception anymore. Mexican citrus are along.
> Best regards from sunny Mexico
I wish it was sunny here. Greetings from a very wet and rainy
There is a slight difference between potatoes and Citrus fruits.
As I see it, if I've understood your message correctly, you are only
looking for the presence or absence of the virus. The situation in
potatoes is complicated by the seed potato classification scheme,
certain levels of virus are allowable in a crop depending on what grade
of seed it is. At the highest grade in which no virus is allowed then I
can see that pooling samples and using PCR/LCR to detect virus would be
preferable to using a less sensitive ELISA, if the costs are comparable.
However at lower grades of seed in which levels af virus are allowable
at 0.5% or higher, how does one quantify a pooled PCR reaction? present
ELISA techniques approach 80-90% accuracy (my figures) and give an
absolute figure on the level infection in the crop.
I'm sure a statistician will tell me that it is possible to
devise a sampling method that could correct for pooled samples, but I
need to be convinced!
I do believe however that as time goes on and the cost of PCR
goes down, it will be possible to test individual samples, though I
think that this is some way off yet.
More information about the Diagnost