Nigeria, Africa's most populous nation, experienced a relatively low COVID-19 burden compared to other regions despite its large population. This low incidence has sparked speculation about underascertainment and other contributing factors, including age structure, mobility, host genetics, environmental factors, and pre-existing immunity. Before the emergence of globally dominant variants like Delta and Omicron, the role of specific genetic diversity in Africa in shaping the observed heterogeneity of incidence and mortality was unclear. This study aimed to characterize the genomic epidemiology of SARS-CoV-2 in Nigeria throughout the pandemic's first three waves, focusing on the emergence of variants of interest (VOIs) B.1.525 and B.1.1.318 and the drivers of transmission in the region. The study highlighted the importance of understanding transmission dynamics in Africa, as findings from other regions might not be directly applicable. Nigeria's high connectivity to neighboring countries and the wider African region, through both air and land travel, suggested a potential role as a significant transmission source. B.1.525 and B.1.1.318 were of particular interest due to experimental evidence suggesting increased infectivity in human and monkey cell lines, and the presence of the E484K mutation known to reduce antibody neutralization.

Several studies have explored the genomic epidemiology of SARS-CoV-2 in Africa. Some research highlighted the importance of incoming travelers in shaping lineage diversity, while others documented the emergence and spread of specific variants in different African countries. These previous studies indicated a need for more comprehensive genomic surveillance to fully understand transmission patterns on the continent. The relatively low COVID-19 incidence in Africa compared to other parts of the world has been a topic of much discussion. Several hypotheses have been proposed, ranging from under-reporting to potential protective factors in the African population. The emergence of variants like B.1.1.7 and the later appearance of Delta and Omicron highlighted the continuous evolution of the virus and raised the importance of comprehensive genomic surveillance across all regions, including under-sampled regions like Africa. The role of travel and human mobility patterns in the spread of variants has been shown to be significant in other regions and was also considered a potentially crucial factor in the spread of SARS-CoV-2 across the African continent.

This study analyzed 1577 SARS-CoV-2 genomes from Nigeria, collected between March 2020 and September 2021 across the first three epidemic waves. Samples were obtained from 25 of Nigeria's 36 states, with uneven sampling across regions and time, especially during the third wave. Pangolin was used to assign lineages to each genome. Phylogenetic Assignment of Named Global Outbreak LINeages (PANGOLIN) v3.1.12 was used to assign lineages and Nextclade v1.3.0 determined global clades and potential primer amplification issues. Lineage dynamics were visualized using Microreact and a GeoJSON file. Maximum likelihood phylogenetic analysis employed IQTREE v2.1.2 with the GTR model and 1000 bootstrap values. TreeTime v0.92 was used for molecular clock quantification and ancestral phylogeny determination. Bayesian phylogenetic analysis included downsampling of 3858 B.1.1.318 sequences and 8278 B.1.525 sequences from GISAID. BEAST v1.10.5 was used with an HKY substitution model, a relaxed molecular clock with a log-normal prior, and an exponential growth coalescent tree prior. Two independent MCMC chains of 200 million states were run. Asymmetric discrete trait analysis in BEAST and a Markov jump counting approach estimated the timing and origin of geographic transitions. Air travel data from the IATA, via BlueDot, quantified international travel volume to and from Nigeria. Google Mobility data provided information on human movement within Nigeria. Introduction Intensity Index (III) and Exportation Intensity Index (EII) were calculated using methods described by Du Plesis et al. Time series of reported deaths from the outbreak.info R package informed calculations. Air travel data was conservatively assumed to be uniform across days.

The study identified more than 35 SARS-CoV-2 lineages circulating during Nigeria's first three epidemic waves. The first wave, initiated two months after the first case detection, ended in August 2020 following a nationwide lockdown. The second wave started in November 2020 after international travel restrictions were lifted. Phylogeographic reconstructions suggested that B.1.1.318 emerged in Africa in early August 2020 and was introduced to Nigeria on at least 53 occasions beginning in November 2020. The majority of introductions originated from other African countries, highlighting Nigeria's strong regional connectivity. Analysis of air travel patterns revealed that introduction risk was primarily driven by regional connectivity during travel restrictions, with the risk increasing significantly after restrictions were lifted. The Introduction Intensity Index (III) showed a peak in introduction risk from Europe and North America, coinciding with their second waves. However, genomic data suggested that most B.1.1.318 introductions came from other African countries, likely underestimating land-based travel. For B.1.525 (Eta), the study estimated its emergence in Nigeria in late July 2020 and found high levels of export to Europe. The Exportation Intensity Index (EII) peaked from December 2020 to February 2021, consistent with the number of estimated exports. However, genomic data underestimated exports to North America compared to the EII, illustrating the impact of uneven global surveillance.

This study demonstrates the importance of integrating genomic, travel, and epidemiological data to understand SARS-CoV-2 transmission. The findings support the emergence of B.1.1.318 and B.1.525 in Africa, with Nigeria playing a significant role in the spread of B.1.525 to Europe. The disparities between genomic estimates and travel-based indices highlight the limitations of genomic data alone when global sampling is uneven. The underestimation of regional connectivity in the import risk index emphasizes the need for comprehensive data collection, including land-based travel, for accurate assessment of transmission dynamics. The relatively low incidence of COVID-19 in Nigeria, even during periods of increased transmission, warrants further research to understand the protective factors that may be at play.

This research underscores the power of integrating diverse data sources to characterize viral emergence and spread, particularly in undersampled regions. The study highlights Nigeria's role as a source of export for the B.1.525 variant, and emphasizes the need for improved genomic surveillance and data-sharing collaborations to gain a more complete understanding of transmission dynamics. Future work should focus on improving data collection methods, including land-based travel data and more comprehensive sampling within Nigeria, especially in under-sampled northern regions. Enhanced international collaboration and investment in genomic surveillance infrastructure in Africa are crucial for effective pandemic preparedness and response.

The study's findings are limited by the uneven sampling across Nigeria, particularly the under-sampling of northern states during the third wave. This uneven sampling limited the accuracy of determining within-country transmission dynamics. The lack of detailed metadata linking samples to travel history or community testing hindered sensitivity analysis. The air travel data did not capture land-based travel, which likely resulted in underestimation of transmission from neighboring countries. The reliance on reported death data to estimate asymptomatic infections resulted in potential biases due to reporting delays and underascertainment in some countries. Overall, the small sampling fraction (0.026% of reported cases) may not be fully representative of the overall viral spread in Nigeria.