A coordinate refers to the location of an amino acid or a nucleotide from the start of a protein or genomic sequence. Hemagglutinin comprises 566 amino acids (aa). The polypeptide is divided into signal peptide (17 aa), HA1 (327 aa) and HA2 (222 aa). All coordinates in the following discussion are in the HA1 frame of reference.
The mutations in the following text are listed in the customary notation. For example, S101N denotes the mutation from S (serine) to N (asparagine) at the coordinate 101 of the HA1 segment.
This research has focused on the HA1 segment because the segment contains the receptor-binding and antigenic domains. Therefore, it is the target of all flu vaccines.
HA1 and HA2 are hinged by an alpha-helical coiled coil in the form a hairpin. The two arms of the hairpin (HA1 and HA2) are held together by two cysteine disulfide bridges. HA1 becomes virulent when the bonds and the hinge are cleaved by the lysosomes inside the host cell.
A/Slovenia/342/2017 and A/Singapore/Infimh-16-0019/2016 were recommended by FutureFlu and WHO/CDC, respectively as the vaccines for the 2018-19 flu season. The BLASTP comparison of the two strains shows that the two vaccines are identical.
Therefore, the vaccine computed by FutureFlu is a perfect match with the one recommended by WHO/CDC.
FutureFlu recommends an A/Slovenia/342/2017-like (SL17) strain for the H3N2 component of the 2018-19 flu vaccine. A/Hong Kong/4801/2014 (HK14) is the current vaccine. The composition of SL17 was calculated with the Fourier-enhanced Phylogeny/Antigenicity Predictive Model (F^P/A-PM) developed by FutureFlu. HK14 and SL17 differ by the mutations N120K, R141G, N170K, I405V, G478E and G483E.
F^P/A-PM shows HK14 has lost efficacy against the prevalent flu strains. In the above phylogenetic tree, HK14 (in red) is on a distant limb of the Clade 3C.2a1, showing the drift of the strains from the vaccine. In contrast, SL17 (in purple) is right in the middle. The computed distance between HK14 and the clade is greater than 1 Antigenic Unit, supporting the conclusion numerically.
Further, the computer models show that A/Michigan/45/2015 (MI15), the current H1N1 component is still effective and should be kept. FutureFlu had recommended an A/Montana/04/2016-like (MT16) vaccine for H1N1 in January 2017. MT16 was a perfect match with MI15 announced by WHO/CDC in March 2017.
BLASTP comparison of A/Montana/04/2016 (MT16, recommended by FutureFlu in January 2017) and A/Michigan/45/2015 (MI15, recommended by WHO in March 2017).
BLASTP shows 100% identities and zero gaps between MT16 and MI15, illustrating that the two vaccines are identical.
CA09: A/California/07/2009 like, current vaccine
MT16: A/Montana/04/2016 like, proposed by FutureFlu
HI Titers computed with the Antigenic Prediction Model:
Clade 6B.1 antigen against CA09 antiserum = 1,142
Clade 6B.1 antigen against MT16 antiserum = 2,616
Autologous titer = 2,560
Antigenic drifts between Clade 6B.1 and
● CA09 = |1 - log2 (1142) / log2 (2560)| = 1.03 AU
● MT16 = |1 - log2 (2616) / log2 (2560)| = 0.03 AU
With an antigenic distance greater than 1 Antigenic Unit, CA09 has drifted from the prevalent strains, while MT16 is highly aligned with Clade 6B.1 (= 0.03 AU). It is also evident in the phylogenetic tree of the clade in the figure above. The vaccine candidates are shown in yellow.