“the economist who thinks he has caught what all the virologists have missed” is right here Paul.
OK now I have to put my mouth where my mouth is. I do honestly think I have answers which are of interest to actual virologists.
Start with the antibody, then find the antigen.
This is really a thought about vaccine development. Now, notably, they seem to be doing that rather well back in my home town (to put it mildly).
“SILVER SPRING, Md. – A series of recently published preclinical study results show that the Spike Ferritin Nanoparticle (SpFN) COVID-19 vaccine developed by researchers at the Walter Reed Army Institute of Research (WRAIR) not only elicits a potent immune response but may also provide broad protection against SARS-CoV-2 variants of concern as well as other coronaviruses.”
But I still have a thought.
The method used to develop all approved vaccines is to take the target pathogen, render it harmless some how, and inject it to evoke an immune response. This has been the approach used in the USA, since it obtained actual independence (during the war of independence then general George Washington immunized soldiers with actual small-pox — the old pre-vaccinia approach called variolation).
The standard approach has yielded miraculous results, but it doesn’t always work. In particular there is no HIV vaccine. There are however monoclonal antibodies which block infection by (some variants) of HIV. An effort to make an antigen which causes production of one such monoclonal didn’t work. The idea was to make the peptide bit of HIV to which it sticks.
I propose something (which I honestly guess has been tried and didn’t work).
FIrst find antibodies which stick to HIV particles. This can be done with display on M13 phages. I say start with imunoglobulin genes from naive b-cells, then mutate the M13s and select. This is a way to get antibodies which require a point mutation to stick — in the immune response there is a high rate of such mutation (part of the program) and then selection of new antibodies which bind tightly to the antigen (this is what happens when you get your booster shot — a long delay first shot to booster can be useful as can take time to get the good mutation (which might involve more than one point mutation) and the new highly sticky antibody to select. Get your 3rd shot (second booster) it will evoke a response roughtly as good against Omicron as the response to two shots is against the original Sars Cov2.
(this also means delay shot to booster longer than 3 or 4 weeks would have been better — I advised doing this — comparison of countries which did what I suggested (eg the UK) and which didn’t (the USA) suggests that I was right (but not for the reason I gave I admit)).
OK now that one has a bunch of antibodies to HIV screen for those which block infection of CD4 cells. This is slow hard work.
All of this is to add to the set of desired antibodies which we would like to produce. As I said some are already known but none blocks all HIV isolates.
Now select peptides which stick to the desired antibodies. (this is phage display not too slow). Now select those which do not stick to other antibodies (don’t want something that sticks to the constant region). again phage display not too slow.
Now add haptens. anti CD3 fab ??? I don’t know. This is pretty established. This gives candidate vaccines
Now screen for vaccines which evoke the desired antibodies. This is super hard slow work using “humice”.
There are other applications for start with the antibody which do not relate to viruses or virology. I think this could be related to developing a useful malaria vaccine, a vaccine for sleeping sickness, and cancer immunotherapy.
I do not defer to credentialed experts.