Re:Coronavirus May ‘Reactivate’ in Cured Patients, Korean CD...
by skwb on Thursday April 09, 2020 @03:04PM
(#22826017)
IMHO I'm fairly skeptical of these claims. There have been a lot of scrambling to get testing online worldwide, so I'm fairly certain that Pr(False Positive) and Pr(False Negative) are reasonably high given circumstances. I don't have the numbers on me right now, but I think sensitivity is only 70%... So it's not hard to imagine a situation where a combination.
The thing that in my mind sorta kills this "reinfection" argument is our previous experience with SARS virus [1]. Figure 1 of the paper shows IgG reactivity for up to 3 years. Anecdotally, I've heard from my immunity bioinformatics colleagues that the spike protein that is recognized by the immune system is very well conserved.
I haven't seen a ton of great evidence that have really pushed my priors against the conclusion that these handful of cases are better explained somehow by faulty testing/some other combination of complication. As always, willing to adjust these priors if given compelling evidence. :)
The thing that in my mind sorta kills this "reinfection" argument is our previous experience with SARS virus [1]. Figure 1 of the paper shows IgG reactivity for up to 3 years. Anecdotally, I've heard from my immunity bioinformatics colleagues that the spike protein that is recognized by the immune system is very well conserved.
I haven't seen a ton of great evidence that have really pushed my priors against the conclusion that these handful of cases are better explained somehow by faulty testing/some other combination of complication. As always, willing to adjust these priors if given compelling evidence. :)
[1] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2851497/