“This NCBI article discusses the matter in much greater detail.”
It seems to me that the article does an analysis of what particular set of COVID features (epitopes) would produce the most thorough immune response - B Cell response, as well as CD4+ and CD8+ T-cell response. They used a bioinformatics (computational analysis) approach, to try to identify which would work best. Kind of an engineering analysis.
“11 selected CD4+ T-cell epitopes were shown to cover 99.36% of global population alleles whereas the 13 CD8+ T-cell epitopes exhibited 91.21% global population coverage” ...”A total of seven linear B-cell epitopes were predicted after filtering based on their antigenicity, surface accessibility, and confirmation using ABCPred” ...”Finally, we have indicated seven multi-epitope sequences where five of the sequences we believe could be integral in the development of an effective peptide-based vaccine that is highly immunogenic and has a wide global population coverage.”
Apparently, this analysis was conducted well after Moderna and Pfizer were already in testing. Those companies may have done their own similar analysis in designing their vaccines, but I did not catch any indication of that in scanning the article, or of how many epitopes those actual vaccines used. I think they used a particular spike protein, how ever many epitopes that might represent.
The article did say that some multi-epitope vaccines had been created (but they may have just been lab experiments, I don’t know - footnotes 44 and 73).
The authors indicated that it might (implied should) be possible to select a set of epitopes to cover a very wide set of viral variations - beyond just COVID-19, possibly including SARS, MERS, or even basically all coronaviruses.
Since this article was published (Oct 2020), I have heard reports that the Army’s/DOD’s Walter Reed medical Center was working to develop such a broadly-based coronavirus vaccine. May or may not be related.