Emergence and transmission dynamics of the FY.4 Omicron variant in Kenya

Abstract

The recombinant FY.4 severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variant was first reported in Kenya in March 2023 and was the dominant circulating variant between April and July 2023. The variant was characterized by two important mutations: Y451H in the receptor-binding domain of the spike protein and P42L in open reading frame 3a. Using phylogenetics and phylodynamic approaches, we investigated the emergence and spread of FY.4 in Kenya and the rest of the world. Our findings suggest FY.4 circulated early in Kenya before its export to North America and Europe. Early circulation of FY.4 in Kenya was predominantly observed in the coastal part of the country, and the estimated time to the most recent common ancestor suggests FY.4 circulated as early as December 2022. The collected genomic and epidemiological data show that the FY.4 variant led to a large local outbreak in Kenya and resulted in localized outbreaks in Europe, North America, and Asia-Pacific. These findings underscore the importance of sustained genomic surveillance, especially in under-sampled regions, in deepening our understanding of the evolution and spread of SARS-CoV-2 variants.

Keywords: FY.4; Kenya; Omicron; phylogenetics.

Figure 1

Circulation of the FY.4 Omicron subvariant. (A) Proportion of SARS-CoV-2 lineages circulating in Kenya from March 2023 to January 2024, the total number of sequences deposited each month in GISAID is shown on top of the stacked bar plot. (B) Global distribution of circulating FY.4 variants between March 2023 and January 2024, the individual colours represent the FY.4 sub-lineages while the circle size denotes the number of genomes deposited in GISAID each month. (C) Time-resolved phylogeny showing the temporal clustering of FY.4 sequences from Kenya and the rest of the world, the colours represent the region from which the sample was collected, while the shape indicates the FY.4 Pango sub-lineages.

Figure 2

(A) Root-to-tip regression plot generated from a TempEst analysis, showing evidence of a temporal signal (R² = 0.326, correlation coefficient = 0.57), the FY.4 sequences are coloured according to the locations (Kenya or Global) and the regression line represents the estimated mean evolutionary rate with error buffers in grey, showing the 90% confidence intervals. (B) The number of transitions was derived from the MCC tree by counting changes between ‘Kenyan’ and ‘other’ locations, transitions were counted when the location of the internal node changed from ‘Kenya’ to ‘Others’ or vice versa or when maintained in the same position. (C) Preliminary discrete trait analysis identified two ancestral clades associated with the spread of FY.4 using the time-resolved tree as the starting tree, the black colour represents the most probable location for Kenyan sequences and background grey the most probable location for global sequences, two ancestral nodes are filled by the plum and salmon colours respectively and the ‘other’ ancestral node occurs primarily on the FY.4.2 sub-lineage, which was not largely observed in Kenya but was predominant in North America.

Figure 3

Dispersal of FY.4 across Kenya over time based on 1000 subsampled trees from a continuous phylogeographic posterior distribution, the nodes of the MCC tree are colour-coded based on the time of occurrence, and the 80% HPD regions are displayed in successive layers with the colours reflecting corresponding time periods for virus spread.

Figure 4

Bayesian phylogeographic reconstruction of FY.4. (A) A Bayesian Skyline plot describing the inferred change in the effective population size of FY.4 infections over time. (B) A time-resolved maximum credibility clade tree with branches coloured by inferred geographic location. (C) A summary of the number of Markov Jumps observed from Kenya to other regions stratified by months.