I'd really like to have a constructive discussion with you - let's try to do it :)
I don't see how, given that you refuse to look at any of the science. I mean, I've asked you to look at the science for each of the claims you make below. I've even spoon fed you some of it. Yet you continue to use the same series of false premises.
Do the primers and probes contain 20 nucleotides? No. They contain a variety of numbers of nucleotides, most more than 20. Many of the papers I linked list the sequences - all you have to do is count. And a single match of 20 nucleotides won't give you a positive test, for the reason I mentioned earlier (a sequence which matches the probe won't get amplified, a sequence which matches a primer won't trigger the probe).
[/QUOTE]
The problem is, I don't work in biology at all, and I'd really like to know which sequence length is used in most of the tests that are out there.
I have read scientists on other forums quote the figure of 20 nucleotides. Also the article I linked to suggested a figure of 18 to 30 nucleotides. I want to emphasise that I am really reporting what others have written, except the calculations of the probabilities which are mine. What figure should I use for the sequence length?
How much do the tests actually vary - I think most people, including politicians, assume there is just one or two COVID tests.
That argument is dead in the water, so why are you still bringing it up?
Well as I say, give me a better figure - preferably with a reference that indicates it relates to a large percentage of the tests in use.
Nobody is claiming it's not possible for other organisms to match, but all the biological entities on the planet are not present in the human mouth or anus.
That is true, but remember I am not just talking about one swab. Most come back negative, but different people will contain different bugs - so at least we should consider all the relevant bugs in the UK, say.
And the human body and its colonizing organisms are the ones we know the most about (that's a guess, but it makes the most sense to me). And again, the probability calculations are vastly more unlikely than a 20 nucleotide match - you need at least 3 matches to generate a "positive" test.
And for many of the tests, you have to get 6 or 9 matches to generate a "positive" test.
You see, the numbers seem to shift a bit, and I'd really like something that I can calculate from. I mean with large numbers like this you really need to calculate. Also, I tell you, a lot of scientists have been effectively shut out of the COVID debate, and that doesn't help. For example, A number of senior scientists ended up posting on an obscure website because they were being 'cancelled' on youtube:
https://off-guardian.org/2020/03/24/12-experts-questioning-the-coronavirus-panic
This is a disgrace, take for example J.P Ioannidis, a professor at Stanford:
https://en.wikipedia.org/wiki/John_Ioannidis
But his contribution was cancelled!
Yes, it would seem that the tests contain multiple tests run in parallel - the number again seems amazingly variable. The only problem with that, is that if each individual test is overwhelmed by the numbers of genomes it encounters, combining the results together may not be much better. Using longer sequences would obviously make more sense - unless there are technical reasons not to do that.
Your claim that they test for false-positives against pure lab samples of various viruses is false. Every paper I've seen on how they test the tests includes tests against clinical samples (I even listed a bunch of actual examples, taken from the papers, in my response to you). I linked to 6 of them, and they were simply the first six links which came up when I googled sars cov 2 rt PCR (or something like that). So again, why are you still sitting on that claim?
Because I don't see how anyone can circumvent Kary Mullis' objection - which is that if you do millions of tests, you will encounter a vast number of different genomes - I don't see how you can fix that.
That's an excellent example of what was mentioned in the ONS article you linked to. Let's assume that all of the positives were false positives. That means that the worst the false-positive rate could be is 0.3%. So later, when using the same test, you are getting positivity rates of 5, 10, 15%, etc., you know that at the very worst, almost all of those tests must be true positives.
True, but by now vast numbers of tests are being performed - so those figures came 'live' from a UK data stream. A 0.3% false positive rate would mean that there are no more new cases of COVID in the UK - we are being locked down for nothing!
I think this is exactly why Kary Mullis said that PCR is not suited for searching for a particular disease.
Who gives a shit? Isn't that the very essence of appeal to authority? Ignore what the considerable science and testing shows. Ignore what all the people with actual expertise in the area are saying about the tests. Instead, hang your hat on the misinterpreted word of a single individual who once won a Nobel prize?
Well I certainly give a shit because he obviously gave a lot of thought as to what you could do with PCR, and presumably, he would have been only too happy to have seen an additional use for it, but he warned against it. He speaks in a strange sort of way, and it took me a while to grasp what he is really saying - but it is fairly obvious when you think about it.
In conclusion, the real problem for both of us, is that the composition of those tests seems so fluid. Discussing or criticising them is like trying to nail jelly to the wall.
Tell me, is there a simple reason why longer sequences are not used as a matter of course? Is it a matter of expense, or what?
Maybe you can answer this. RT-PCR obviously starts by converting the RNA to DNA, but does the extraneous DNA in the sample stay in at that point, or can it be removed? Obviously, if it is removed, much of my argument would disappear.
David