1) Get samples of the three main vaxxes and sequence the rna (dna for JnJ)
2) within the same vax, calculate Jaccard similarity of all the sequences you found.
> what’s the variation like within batches and what’s the variation between batch to batch based just on the sequences you measured?
Ok then:
3) compare that to what the known nucleotide sequences SHOULD be to encode for a spike protein (which is known)
> how close did they each get?
4) shoot the vax into mammalian cells in a Petri dish and allow them to express whatever the vaxxes are coding for. Harvest the proteins and analyze them with X-ray crystallography and NMR.
> how close are the proteins — as expressed by each company’s nucleotide sequence — match up to the known spike protein structure?
Bonus round:
5) take the known spike protein structure and do some folding prediction to see if, with a little bit of nucleotide sequence tweaking, you can get a folded protein that is very very nasty.
If you can do that:
> compare what you found about the variation between companies, batches, and within batches to see if it’s plausible that the very nasty protein could get expressed in real people.
You know why you haven’t seen this study, I’m sure.
Who’s got the balls to do it?
2) within the same vax, calculate Jaccard similarity of all the sequences you found.
> what’s the variation like within batches and what’s the variation between batch to batch based just on the sequences you measured?
Ok then:
3) compare that to what the known nucleotide sequences SHOULD be to encode for a spike protein (which is known)
> how close did they each get?
4) shoot the vax into mammalian cells in a Petri dish and allow them to express whatever the vaxxes are coding for. Harvest the proteins and analyze them with X-ray crystallography and NMR.
> how close are the proteins — as expressed by each company’s nucleotide sequence — match up to the known spike protein structure?
Bonus round:
5) take the known spike protein structure and do some folding prediction to see if, with a little bit of nucleotide sequence tweaking, you can get a folded protein that is very very nasty.
If you can do that:
> compare what you found about the variation between companies, batches, and within batches to see if it’s plausible that the very nasty protein could get expressed in real people.
You know why you haven’t seen this study, I’m sure.
Who’s got the balls to do it?