The state of genome sequences available publicly from WHO AFRO countries

Overall, there has been a much-increased global effort and speed in pathogen sequencing and availing the data in public depositories during this SARS CoV-2 pandemic. Average lag between sample collection and sequence availability has been estimated to be 3 months. We have calculated that for other previous human infective viruses, there was an average of over two and a half years, between sample collection and public sequence availability in 2015. In 2019, this had reduced to an average of about 1 year.

During the SARS-CoV-2 pandemic, several African countries have published in-country sequences faster than the average 3 months. For example, 3 days after the confirmation of Nigeria’s first COVID-19 case, the genome sequencing results of the SARS-CoV-2 specimen from the country were announced on 1 March 2020.7 However, to date, there have been no published analyses of SARS-CoV-2 genomes at a continent level in Africa to summarise the patterns on the continent. We have therefore conducted an analysis of all the publicly available sequences from the WHO Africa Region member countries to summarise the currently available information from SARS-CoV-2 genomes.

Data comprising all SARS-CoV-2 genome sequences and the corresponding metadata from WHO Africa region (AFRO) on the GISAID sequence data repository were downloaded on 30 November 2020 (https://www.gisaid.org). Spatiotemporal phylogenetic analyses were conducted using the complete genomes available at the time, from African countries and other continents. Over representative sequences were subsampled. The number of introductions to WHO AFRO member countries were estimated by inferring the ancestral states on the phylogenetic tree using the likelihood reconstruction method.

As of 30 November 2020, 3995 SARS-CoV-2 genomes had been submitted to GISAID from 23 countries in the WHO Africa Region (out of 47). These represented ~2% of publicly available sequences globally. The majority of genomes were from South Africa (56%), Democratic Republic of the Congo (9%) and Gambia (9%) (figure 1A,B).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Summary of sequence data for WHO AFRO countries. (A) Heat map of SARS-CoV-2 genomes available on GISAID by African country. (B) Comparison of cumulative reported cases and number of sequences available per country, with the variables on the axes, number of sequences and cumulative cases log transformed. Different colours represent different countries. (C) Global lineages in African countries (colours present different African countries as in B) compared with other regions in the world (in grey). (D) Genomes from individual countries highlighted on approximate time-scaled global subsampled tree (n=3266), with tips coloured by the sampling locations as per country-colour key in (C). Lineage B.1 is indicated on the node. (E) SARS-CoV-2 genomes available on GISAID by collection date. Colours for WHO AFRO countries are consistent in (B) to (E), other world regions are in grey.

SARS Cov-2 in WHO Africa countries

Eighty-four lineages were identified in the sequences from WHO AFRO countries. Currently, 266 lineages (including A, B, C and D lineages and all their descendent sublineages) that have so far contributed to active spread globally have been identified by the PANGOLIN phylogenetic classification.8 Knowledge of circulating lineages is important for multiple reasons, including determining potential efficacy of vaccines, as has been demonstrated by the challenges for vaccine development for other viruses such as HIV-1, influenza or Dengue driven by viral diversity. A study analysing 18 514 SARS-CoV-2 sequences sampled since December 2019 concluded that although there are some reported differences in SARS-CoV-2 genomes, viral diversity is not sufficient to reduce the efficacy of vaccines currently under development.9

As of 30 November, the most prevalent lineage globally was B.1 (figure 1C, D). Our analyses show that 86% of all African SARS-CoV-2 genomes belong to this B.1 lineage which represents a large European lineage that corresponds to the Italian outbreak in February, and its descendant sublineages. In South Africa, B.1.106 was identified in April 2020 as the dominant sublineage, but reports now indicate that this lineage became extinct after the effective control of nosocomial outbreaks.10 These include the novel C.1 lineage, which has 16 mutations compared with the original Wuhan sequence. The C.1 lineage has been reported as the most geographically widespread lineage in South Africa.10 The implications of these changes, if any, in terms of transmission and disease outcome are still to be determined.

Our analyses indicate that there have been multiple separate introductions of SARS-CoV-2 into Africa. We estimated that based on currently available sequences, there have been at least 211 separate introductions of the virus into WHO AFRO countries from other continents and African countries outside the WHO AFRO region (figure 1D). Of these, 43% were introduced from Europe, supporting the epidemiological case tracing findings. Most of these introductions do not appear to have spread between countries in Africa. However, 80 introductions from within Africa were tentatively identified, but the small number of sequences currently available from the African countries means that directionality of these introductions cannot as yet, be determined.

What still needs to be done

There is still need for sequencing of the virus in other African countries and for the sequences to be deposited in publicly available data bases. Our analyses indicate that we are still missing sequences from 24 WHO AFRO countries and, furthermore, some of the countries with sequence data have submitted a total of less than 10 sequences to date. The launch of a network of laboratories in September 2020 to reinforce genome sequencing of SARS-CoV-2 in Africa by the WHO World and the Africa Centres for Disease Control and Prevention (Africa CDC) is a welcome development.11 This strengthens the capacity of the member states already provided by other partners on the continent including the African Center of Excellence for Genomics of Infectious Diseases (ACEGID), Nigerian Center for Disease Control (Nigeria CDC) and Tackling Infections to Benefit Africa (TIBA).12

There is also need to ensure that the sequencing is science-led or purpose-led so that limited resources are not wasted on sequencing for the sake of sequencing. Finally, there is need to ensure that supporting data are collected and well curated so that patient meta-data including clinical, immunological and prognosis can easily be related to the sequence data. This will ensure that the impact of molecular changes on phenotypic attributes can be more readily identified. For example, when SARS-CoV-2 emerged in Wuhan, it had a D residue (aspartate) at position 614 in the sequence of the virus’s spike protein. By June, the D residue had been replaced by G (glycine) on most continents. This mutation is thought to be associated with higher infectivity but lower disease severity.6 13 Our analyses have shown that 94% of African SARS-CoV-2 genomes have the D614G mutation in spike protein.