“.....What do we mean when we say somebody has ‘tested positive’...
“perhaps some kind FReeper would write down what the answer is.”
There’s a lot of excess verbiage in that article, so if one doesn’t read it all (or doesn’t really much understand it all) it’s easy to miss - but near the end one finds that’s the point of the article: we don’t really know what a positive test is.
I don’t know how much of the article you were able to follow, but here are some of the key things. We don’t really test for the virus per se; we test for segments of genetic material that indicate the presence of segments of proteins (supposedly) unique to the virus.
In order to detect that genetic material, we have to increase the amount present - we do that by a form of (for want of a better word) “molecular Xeroxing” called PCR. Every time you run a cycle in PCR, you double the amount of material present - for another crude analogy, recall the movie “Blow Up” where a guy kept increasing the size of a portion of a photograph until he could detect a particular detail.
For the COVID test, they have to do somewhere between 30 and 40 “doublings” in order to say they’ve detected evidence for the suspect proteins (and hence, presumably, the virus).
Problem #1: it doesn’t seem to be clear how many doublings are a reliable indicator for a positive test, and moreover, whether everybody is using the same standard - what might be a negative result in country A could be a positive result in country B.
Problem #2: it’s not clear why any particular number of doublings should be the threshold for a positive test. Why should no detectable material after 34 doublings be negative, and some detectable material after 35 doublings be positive?
Problem #3: You have to accept that the nucleic acid segments targeted as indicators of COVID-19 proteins really are unique to that virus and that virus only. (I have no idea how reasonable an assumption that is - but they are targeting only a piece of the protein structure, not the entire thing, and lots of these viruses have somewhat related proteins).
Problem #4: unstated in the article, but well-known in the business - PCR is really sensitive to accidental contamination - a little speck on you “original photograph” can become a giant blob if you enlarge it enough times.
Wow. Excellent breakdown. You must be a teacher, you made it all so clear. Thank you.
Sounds its kind of like, but different from growing bacteria to determine what kind of urinary infection it is since you are working with invisible parts of virons instead of much larger visible (under a microscope) bacteria?