Seamless, rapid, and accurate analyses of outbreak genomic data using split k-mer analysis.

Sequence variation observed in populations of pathogens can be used for important public health and evolutionary genomic analyses, especially outbreak analysis and transmission reconstruction. Identifying this variation is typically achieved by aligning sequence reads to a reference genome, but this approach is susceptible to reference biases and requires careful filtering of called genotypes. There is a need for tools that can process this growing volume of bacterial genome data, providing rapid results, but that remain simple so they can be used without highly trained bioinformaticians, expensive data analysis, and long-term storage and processing of large files. Here we describe split k-mer analysis (SKA2), a method that supports both reference-free and reference-based mapping to quickly and accurately genotype populations of bacteria using sequencing reads or genome assemblies. SKA2 is highly accurate for closely related samples, and in outbreak simulations, we show superior variant recall compared with reference-based methods, with no false positives. SKA2 can also accurately map variants to a reference and be used with recombination detection methods to rapidly reconstruct vertical evolutionary history. SKA2 is many times faster than comparable methods and can be used to add new genomes to an existing call set, allowing sequential use without the need to reanalyze entire collections. With an inherent absence of reference bias, high accuracy, and a robust implementation, SKA2 has the potential to become the tool of choice for genotyping bacteria. SKA2 is implemented in Rust and is freely available as open-source software.

CELEBRIMBOR: core and accessory genes from metagenomes.

MOTIVATION: Metagenome-Assembled Genomes (MAGs) or Single-cell Amplified Genomes (SAGs) are often incomplete, with sequences missing due to errors in assembly or low coverage. This presents a particular challenge for the identification of true gene frequencies within a microbial population, as core genes missing in only a few assemblies will be mischaracterized by current pangenome approaches. RESULTS: Here, we present CELEBRIMBOR, a Snakemake pangenome analysis pipeline which uses a measure of genome completeness to automatically adjust the frequency threshold at which core genes are identified, enabling accurate core gene identification in MAGs and SAGs. AVAILABILITY AND IMPLEMENTATION: CELEBRIMBOR is published under open source Apache 2.0 licence at https://github.com/bacpop/CELEBRIMBOR and is available as a Docker container from this repository. Supplementary material is available in the online version of the article.

Comparison of gene-by-gene and genome-wide short nucleotide sequence-based approaches to define the global population structure of Streptococcus pneumoniae.

Defining the population structure of a pathogen is a key part of epidemiology, as genomically related isolates are likely to share key clinical features such as antimicrobial resistance profiles and invasiveness. Multiple different methods are currently used to cluster together closely related genomes, potentially leading to inconsistency between studies. Here, we use a global dataset of 26 306 Streptococcus pneumoniae genomes to compare four clustering methods: gene-by-gene seven-locus MLST, core genome MLST (cgMLST)-based hierarchical clustering (HierCC) assignments, life identification number (LIN) barcoding and k-mer-based PopPUNK clustering (known as GPSCs in this species). We compare the clustering results with phylogenetic and pan-genome analyses to assess their relationship with genome diversity and evolution, as we would expect a good clustering method to form a single monophyletic cluster that has high within-cluster similarity of genomic content. We show that the four methods are generally able to accurately reflect the population structure based on these metrics and that the methods were broadly consistent with each other. We investigated further to study the discrepancies in clusters. The greatest concordance was seen between LIN barcoding and HierCC (adjusted mutual information score=0.950), which was expected given that both methods utilize cgMLST, but have different methods for defining an individual cluster and different core genome schema. However, the existence of differences between the two methods shows that the selection of a core genome schema can introduce inconsistencies between studies. GPSC and HierCC assignments were also highly concordant (AMI=0.946), showing that k-mer-based methods which use the whole genome and do not require the careful selection of a core genome schema are just as effective at representing the population structure. Additionally, where there were differences in clustering between these methods, this could be explained by differences in the accessory genome that were not identified in cgMLST. We conclude that for S. pneumoniae, standardized and stable nomenclature is important as the number of genomes available expands. Furthermore, the research community should transition away from seven-locus MLST, whilst cgMLST, GPSC and LIN assignments should be used more widely. However, to allow for easy comparison between studies and to make previous literature relevant, the reporting of multiple clustering names should be standardized within the research.

Pangenome-spanning epistasis and coselection analysis via de Bruijn graphs.

Studies of bacterial adaptation and evolution are hampered by the difficulty of measuring traits such as virulence, drug resistance, and transmissibility in large populations. In contrast, it is now feasible to obtain high-quality complete assemblies of many bacterial genomes thanks to scalable high-accuracy long-read sequencing technologies. To exploit this opportunity, we introduce a phenotype- and alignment-free method for discovering coselected and epistatically interacting genomic variation from genome assemblies covering both core and accessory parts of genomes. Our approach uses a compact colored de Bruijn graph to approximate the intragenome distances between pairs of loci for a collection of bacterial genomes to account for the impacts of linkage disequilibrium (LD). We demonstrate the versatility of our approach to efficiently identify associations between loci linked with drug resistance and adaptation to the hospital niche in the major human bacterial pathogens Streptococcus pneumoniae and Enterococcus faecalis.

Recent approaches in computational modelling for controlling pathogen threats.

In this review, we assess the status of computational modelling of pathogens. We focus on three disparate but interlinked research areas that produce models with very different spatial and temporal scope. First, we examine antimicrobial resistance (AMR). Many mechanisms of AMR are not well understood. As a result, it is hard to measure the current incidence of AMR, predict the future incidence, and design strategies to preserve existing antibiotic effectiveness. Next, we look at how to choose the finite number of bacterial strains that can be included in a vaccine. To do this, we need to understand what happens to vaccine and non-vaccine strains after vaccination programmes. Finally, we look at within-host modelling of antibody dynamics. The SARS-CoV-2 pandemic produced huge amounts of antibody data, prompting improvements in this area of modelling. We finish by discussing the challenges that persist in understanding these complex biological systems.

Inter-species gene flow drives ongoing evolution of Streptococcus pyogenes and Streptococcus dysgalactiae subsp. equisimilis.

Streptococcus dysgalactiae subsp. equisimilis (SDSE) is an emerging cause of human infection with invasive disease incidence and clinical manifestations comparable to the closely related species, Streptococcus pyogenes. Through systematic genomic analyses of 501 disseminated SDSE strains, we demonstrate extensive overlap between the genomes of SDSE and S. pyogenes. More than 75% of core genes are shared between the two species with one third demonstrating evidence of cross-species recombination. Twenty-five percent of mobile genetic element (MGE) clusters and 16 of 55 SDSE MGE insertion regions were shared across species. Assessing potential cross-protection from leading S. pyogenes vaccine candidates on SDSE, 12/34 preclinical vaccine antigen genes were shown to be present in >99% of isolates of both species. Relevant to possible vaccine evasion, six vaccine candidate genes demonstrated evidence of inter-species recombination. These findings demonstrate previously unappreciated levels of genomic overlap between these closely related pathogens with implications for streptococcal pathobiology, disease surveillance and prevention.

Optimising machine learning prediction of minimum inhibitory concentrations in Klebsiella pneumoniae.

Minimum Inhibitory Concentrations (MICs) are the gold standard for quantitatively measuring antibiotic resistance. However, lab-based MIC determination can be time-consuming and suffers from low reproducibility, and interpretation as sensitive or resistant relies on guidelines which change over time. Genome sequencing and machine learning promise to allow in silico MIC prediction as an alternative approach which overcomes some of these difficulties, albeit the interpretation of MIC is still needed. Nevertheless, precisely how we should handle MIC data when dealing with predictive models remains unclear, since they are measured semi-quantitatively, with varying resolution, and are typically also left- and right-censored within varying ranges. We therefore investigated genome-based prediction of MICs in the pathogen Klebsiella pneumoniae using 4367 genomes with both simulated semi-quantitative traits and real MICs. As we were focused on clinical interpretation, we used interpretable rather than black-box machine learning models, namely, Elastic Net, Random Forests, and linear mixed models. Simulated traits were generated accounting for oligogenic, polygenic, and homoplastic genetic effects with different levels of heritability. Then we assessed how model prediction accuracy was affected when MICs were framed as regression and classification. Our results showed that treating the MICs differently depending on the number of concentration levels of antibiotic available was the most promising learning strategy. Specifically, to optimise both prediction accuracy and inference of the correct causal variants, we recommend considering the MICs as continuous and framing the learning problem as a regression when the number of observed antibiotic concentration levels is large, whereas with a smaller number of concentration levels they should be treated as a categorical variable and the learning problem should be framed as a classification. Our findings also underline how predictive models can be improved when prior biological knowledge is taken into account, due to the varying genetic architecture of each antibiotic resistance trait. Finally, we emphasise that incrementing the population database is pivotal for the future clinical implementation of these models to support routine machine-learning based diagnostics.

Impact of vaccinations, boosters and lockdowns on COVID-19 waves in French Polynesia.

Estimating the impact of vaccination and non-pharmaceutical interventions on COVID-19 incidence is complicated by several factors, including successive emergence of SARS-CoV-2 variants of concern and changing population immunity from vaccination and infection. We develop an age-structured multi-strain COVID-19 transmission model and inference framework to estimate vaccination and non-pharmaceutical intervention impact accounting for these factors. We apply this framework to COVID-19 waves in French Polynesia and estimate that the vaccination programme averted 34.8% (95% credible interval: 34.5-35.2%) of 223,000 symptomatic cases, 49.6% (48.7-50.5%) of 5830 hospitalisations and 64.2% (63.1-65.3%) of 1540 hospital deaths that would have occurred in a scenario without vaccination up to May 2022. We estimate the booster campaign contributed 4.5%, 1.9%, and 0.4% to overall reductions in cases, hospitalisations, and deaths. Our results suggest that removing lockdowns during the first two waves would have had non-linear effects on incidence by altering accumulation of population immunity. Our estimates of vaccination and booster impact differ from those for other countries due to differences in age structure, previous exposure levels and timing of variant introduction relative to vaccination, emphasising the importance of detailed analysis that accounts for these factors.

Accurate and fast graph-based pangenome annotation and clustering with ggCaller.

Bacterial genomes differ in both gene content and sequence mutations, which underlie extensive phenotypic diversity, including variation in susceptibility to antimicrobials or vaccine-induced immunity. To identify and quantify important variants, all genes within a population must be predicted, functionally annotated, and clustered, representing the "pangenome." Despite the volume of genome data available, gene prediction and annotation are currently conducted in isolation on individual genomes, which is computationally inefficient and frequently inconsistent across genomes. Here, we introduce the open-source software graph-gene-caller (ggCaller). ggCaller combines gene prediction, functional annotation, and clustering into a single workflow using population-wide de Bruijn graphs, removing redundancy in gene annotation and resulting in more accurate gene predictions and orthologue clustering. We applied ggCaller to simulated and real-world bacterial data sets containing hundreds or thousands of genomes, comparing it to current state-of-the-art tools. ggCaller has considerable speed-ups with equivalent or greater accuracy, particularly with data sets containing complex sources of error, such as assembly contamination or fragmentation. ggCaller is also an important extension to bacterial genome-wide association studies, enabling querying of annotated graphs for functional analyses. We highlight this application by functionally annotating DNA sequences with significant associations to tetracycline and macrolide resistance in Streptococcus pneumoniae, identifying key resistance determinants that were missed when using only a single reference genome. ggCaller is a novel bacterial genome analysis tool with applications in bacterial evolution and epidemiology.

Genealogical inference and more flexible sequence clustering using iterative-PopPUNK.

Bacterial genome data are accumulating at an unprecedented speed due to the routine use of sequencing in clinical diagnoses, public health surveillance, and population genetics studies. Genealogical reconstruction is fundamental to many of these uses; however, inferring genealogy from large-scale genome data sets quickly, accurately, and flexibly is still a challenge. Here, we extend an alignment- and annotation-free method, PopPUNK, to increase its flexibility and interpretability across data sets. Our method, iterative-PopPUNK, rapidly produces multiple consistent cluster assignments across a range of sequence identities. By constructing a partially resolved genealogical tree with respect to these clusters, users can select a resolution most appropriate for their needs. We showed the accuracy of clusters at all levels of similarity and genealogical inference of iterative-PopPUNK based on simulated data and obtained phylogenetically concordant results in real data sets from seven bacterial species. Using two example sets of Escherichia/Shigella and Vibrio parahaemolyticus genomes, we show that iterative-PopPUNK can achieve cluster resolutions ranging from phylogroup down to sequence typing (ST). The iterative-PopPUNK algorithm is implemented in the "PopPUNK_iterate" program, available as part of the PopPUNK package.

Inferring the heritability of bacterial traits in the era of machine learning.

Quantification of heritability is a fundamental desideratum in genetics, which allows an assessment of the contribution of additive genetic variation to the variability of a trait of interest. The traditional computational approaches for assessing the heritability of a trait have been developed in the field of quantitative genetics. However, the rise of modern population genomics with large sample sizes has led to the development of several new machine learning-based approaches to inferring heritability. In this article, we systematically summarize recent advances in machine learning which can be used to infer heritability. We focus on an application of these methods to bacterial genomes, where heritability plays a key role in understanding phenotypes such as antibiotic resistance and virulence, which are particularly important due to the rising frequency of antimicrobial resistance. By designing a heritability model incorporating realistic patterns of genome-wide linkage disequilibrium for a frequently recombining bacterial pathogen, we test the performance of a wide spectrum of different inference methods, including also GCTA. In addition to the synthetic data benchmark, we present a comparison of the methods for antibiotic resistance traits for multiple bacterial pathogens. Insights from the benchmarking and real data analyses indicate a highly variable performance of the different methods and suggest that heritability inference would likely benefit from tailoring of the methods to the specific genetic architecture of the target organism. Availability and implementation: The R codes and data used in the numerical experiments are available at: https://github.com/tienmt/her_MLs.

Dissemination of carbapenemase-producing Enterobacterales in Ireland from 2012 to 2017: a retrospective genomic surveillance study.

The spread of carbapenemase-producing Enterobacterales (CPE) is of major public health concern. The transmission dynamics of CPE in hospitals, particularly at the national level, are not well understood. Here, we describe a retrospective nationwide genomic surveillance study of CPE in Ireland between 2012 and 2017. We sequenced 746 national surveillance CPE samples obtained between 2012 and 2017. After clustering the sequences, we used thresholds based on pairwise SNPs, and reported within-host diversity along with epidemiological data to infer recent putative transmissions. All clusters in circulating clones, derived from high-resolution phylogenies, of a species (Klebsiella pneumoniae, Escherichia coli, Klebsiella oxytoca, Enterobacter cloacae, Enterobacter hormaechei and Citrobacter freundii) were individually examined for evidence of transmission. Antimicrobial resistance trends over time were also assessed. We identified 352 putative transmission events in six species including widespread and frequent transmissions in three species. We detected putative outbreaks in 4/6 species with three hospitals experiencing prolonged outbreaks. The bla OXA-48 gene was the main cause of carbapenem resistance in Ireland in almost all species. An expansion in the number of sequence types carrying bla OXA-48 was an additional cause of the increasing prevalence of carbapenemase-producing K. pneumoniae and E. coli.

BlpC-mediated selfish program leads to rapid loss of Streptococcus pneumoniae clonal diversity during infection.

Successful colonization of a host requires bacterial adaptation through genetic and population changes that are incompletely defined. Using chromosomal barcoding and high-throughput sequencing, we investigate the population dynamics of Streptococcus pneumoniae during infant mouse colonization. Within 1 day post inoculation, diversity was reduced >35-fold with expansion of a single clonal lineage. This loss of diversity was not due to immune factors, microbiota, or exclusive genetic drift. Rather, bacteriocins induced by the BlpC-quorum sensing pheromone resulted in predation of kin cells. In this intra-strain competition, the subpopulation reaching a quorum likely eliminates others that have yet to activate the blp locus. Additionally, this reduced diversity restricts the number of unique clones that establish colonization during transmission between hosts. Genetic variation in the blp locus was also associated with altered transmissibility in a human population, further underscoring the importance of BlpC in clonal selection and its role as a selfish element.

Mandrake: visualizing microbial population structure by embedding millions of genomes into a low-dimensional representation.

In less than a decade, population genomics of microbes has progressed from the effort of sequencing dozens of strains to thousands, or even tens of thousands of strains in a single study. There are now hundreds of thousands of genomes available even for a single bacterial species, and the number of genomes is expected to continue to increase at an accelerated pace given the advances in sequencing technology and widespread genomic surveillance initiatives. This explosion of data calls for innovative methods to enable rapid exploration of the structure of a population based on different data modalities, such as multiple sequence alignments, assemblies and estimates of gene content across different genomes. Here, we present Mandrake, an efficient implementation of a dimensional reduction method tailored for the needs of large-scale population genomics. Mandrake is capable of visualizing population structure from millions of whole genomes, and we illustrate its usefulness with several datasets representing major pathogens. Our method is freely available both as an analysis pipeline (https://github.com/johnlees/mandrake) and as a browser-based interactive application (https://gtonkinhill.github.io/mandrake-web/). This article is part of a discussion meeting issue 'Genomic population structures of microbial pathogens'.

Pneumococcal genetic variability in age-dependent bacterial carriage.

The characteristics of pneumococcal carriage vary between infants and adults. Host immune factors have been shown to contribute to these age-specific differences, but the role of pathogen sequence variation is currently less well-known. Identification of age-associated pathogen genetic factors could leadto improved vaccine formulations. We therefore performed genome sequencing in a large carriage cohort of children and adults and combined this with data from an existing age-stratified carriage study. We compiled a dictionary of pathogen genetic variation, including serotype, strain, sequence elements, single-nucleotide polymorphisms (SNPs), and clusters of orthologous genes (COGs) for each cohort - all of which were used in a genome-wide association with host age. Age-dependent colonization showed weak evidence of being heritable in the first cohort (h2 = 0.10, 95% CI 0.00-0.69) and stronger evidence in the second cohort (h2 = 0.56, 95% CI 0.23-0.87). We found that serotypes and genetic background (strain) explained a proportion of the heritability in the first cohort (h2serotype = 0.07, 95% CI 0.04-0.14 and h2GPSC = 0.06, 95% CI 0.03-0.13) and the second cohort (h2serotype = 0.11, 95% CI 0.05-0.21 and h2GPSC = 0.20, 95% CI 0.12-0.31). In a meta-analysis of these cohorts, we found one candidate association (p=1.2 × 10-9) upstream of an accessory Sec-dependent serine-rich glycoprotein adhesin. Overall, while we did find a small effect of pathogen genome variation on pneumococcal carriage between child and adult hosts, this was variable between populations and does not appear to be caused by strong effects of individual genes. This supports proposals for adaptive future vaccination strategies that are primarily targeted at dominant circulating serotypes and tailored to the composition of the pathogen populations.

Targeted control of pneumolysin production by a mobile genetic element in Streptococcus pneumoniae.

Streptococcus pneumoniae is a major human pathogen that can cause severe invasive diseases such as pneumonia, septicaemia and meningitis. Young children are at a particularly high risk, with an estimated 3-4 million cases of severe disease and between 300 000 and 500 000 deaths attributable to pneumococcal disease each year. The haemolytic toxin pneumolysin (Ply) is a primary virulence factor for this bacterium, yet despite its key role in pathogenesis, immune evasion and transmission, the regulation of Ply production is not well defined. Using a genome-wide association approach, we identified a large number of potential affectors of Ply activity, including a gene acquired horizontally on the antibiotic resistance-conferring Integrative and Conjugative Element (ICE) ICESp23FST81. This gene encodes a novel modular protein, ZomB, which has an N-terminal UvrD-like helicase domain followed by two Cas4-like domains with potent ATP-dependent nuclease activity. We found the regulatory effect of ZomB to be specific for the ply operon, potentially mediated by its high affinity for the BOX repeats encoded therein. Using a murine model of pneumococcal colonization, we further demonstrate that a ZomB mutant strain colonizes both the upper respiratory tract and lungs at higher levels when compared to the wild-type strain. While the antibiotic resistance-conferring aspects of ICESp23FST81 are often credited with contributing to the success of the S. pneumoniae lineages that acquire it, its ability to control the expression of a major virulence factor implicated in bacterial transmission is also likely to have played an important role.

Pneumococcal capsule blocks protection by immunization with conserved surface proteins.

Vaccines targeting Streptococcus pneumoniae (Spn) are limited by dependence on capsular polysaccharide and its serotype diversity. More broadly-based approaches using common protein antigens have not resulted in a licensed vaccine. Herein, we used an unbiased, genome-wide approach to find novel vaccine antigens to disrupt carriage modeled in mice. A Tn-Seq screen identified 198 genes required for colonization of which 16 are known to express conserved, immunogenic surface proteins. After testing defined mutants for impaired colonization of infant and adult mice, 5 validated candidates (StkP, PenA/Pbp2a, PgdA, HtrA, and LytD/Pce/CbpE) were used as immunogens. Despite induction of antibody recognizing the Spn cell surface, there was no protection against Spn colonization. There was, however, protection against an unencapsulated Spn mutant. This result correlated with increased antibody binding to the bacterial surface in the absence of capsule. Our findings demonstrate how the pneumococcal capsule interferes with mucosal protection by antibody to common protein targets.

Non-pharmaceutical interventions, vaccination, and the SARS-CoV-2 delta variant in England: a mathematical modelling study.

BACKGROUND: England's COVID-19 roadmap out of lockdown policy set out the timeline and conditions for the stepwise lifting of non-pharmaceutical interventions (NPIs) as vaccination roll-out continued, with step one starting on March 8, 2021. In this study, we assess the roadmap, the impact of the delta (B.1.617.2) variant of SARS-CoV-2, and potential future epidemic trajectories. METHODS: This mathematical modelling study was done to assess the UK Government's four-step process to easing lockdown restrictions in England, UK. We extended a previously described model of SARS-CoV-2 transmission to incorporate vaccination and multi-strain dynamics to explicitly capture the emergence of the delta variant. We calibrated the model to English surveillance data, including hospital admissions, hospital occupancy, seroprevalence data, and population-level PCR testing data using a Bayesian evidence synthesis framework, then modelled the potential trajectory of the epidemic for a range of different schedules for relaxing NPIs. We estimated the resulting number of daily infections and hospital admissions, and daily and cumulative deaths. Three scenarios spanning a range of optimistic to pessimistic vaccine effectiveness, waning natural immunity, and cross-protection from previous infections were investigated. We also considered three levels of mixing after the lifting of restrictions. FINDINGS: The roadmap policy was successful in offsetting the increased transmission resulting from lifting NPIs starting on March 8, 2021, with increasing population immunity through vaccination. However, because of the emergence of the delta variant, with an estimated transmission advantage of 76% (95% credible interval [95% CrI] 69-83) over alpha, fully lifting NPIs on June 21, 2021, as originally planned might have led to 3900 (95% CrI 1500-5700) peak daily hospital admissions under our central parameter scenario. Delaying until July 19, 2021, reduced peak hospital admissions by three fold to 1400 (95% CrI 700-1700) per day. There was substantial uncertainty in the epidemic trajectory, with particular sensitivity to the transmissibility of delta, level of mixing, and estimates of vaccine effectiveness. INTERPRETATION: Our findings show that the risk of a large wave of COVID-19 hospital admissions resulting from lifting NPIs can be substantially mitigated if the timing of NPI relaxation is carefully balanced against vaccination coverage. However, with the delta variant, it might not be possible to fully lift NPIs without a third wave of hospital admissions and deaths, even if vaccination coverage is high. Variants of concern, their transmissibility, vaccine uptake, and vaccine effectiveness must be carefully monitored as countries relax pandemic control measures. FUNDING: National Institute for Health Research, UK Medical Research Council, Wellcome Trust, and UK Foreign, Commonwealth and Development Office.

Key epidemiological drivers and impact of interventions in the 2020 SARS-CoV-2 epidemic in England.

We fitted a model of SARS-CoV-2 transmission in care homes and the community to regional surveillance data for England. Compared with other approaches, our model provides a synthesis of multiple surveillance data streams into a single coherent modeling framework, allowing transmission and severity to be disentangled from features of the surveillance system. Of the control measures implemented, only national lockdown brought the reproduction number (Rt eff) below 1 consistently; if introduced 1 week earlier, it could have reduced deaths in the first wave from an estimated 48,600 to 25,600 [95% credible interval (CrI): 15,900 to 38,400]. The infection fatality ratio decreased from 1.00% (95% CrI: 0.85 to 1.21%) to 0.79% (95% CrI: 0.63 to 0.99%), suggesting improved clinical care. The infection fatality ratio was higher in the elderly residing in care homes (23.3%, 95% CrI: 14.7 to 35.2%) than those residing in the community (7.9%, 95% CrI: 5.9 to 10.3%). On 2 December 2020, England was still far from herd immunity, with regional cumulative infection incidence between 7.6% (95% CrI: 5.4 to 10.2%) and 22.3% (95% CrI: 19.4 to 25.4%) of the population. Therefore, any vaccination campaign will need to achieve high coverage and a high degree of protection in vaccinated individuals to allow nonpharmaceutical interventions to be lifted without a resurgence of transmission.

Within-country age-based prioritisation, global allocation, and public health impact of a vaccine against SARS-CoV-2: A mathematical modelling analysis.

The worldwide endeavour to develop safe and effective COVID-19 vaccines has been extraordinary, and vaccination is now underway in many countries. However, the doses available in 2021 are likely to be limited. We extend a mathematical model of SARS-CoV-2 transmission across different country settings to evaluate the public health impact of potential vaccines using WHO-developed target product profiles. We identify optimal vaccine allocation strategies within- and between-countries to maximise averted deaths under constraints on dose supply. We find that the health impact of SARS-CoV-2 vaccination depends on the cumulative population-level infection incidence when vaccination begins, the duration of natural immunity, the trajectory of the epidemic prior to vaccination, and the level of healthcare available to effectively treat those with disease. Within a country we find that for a limited supply (doses for < 20% of the population) the optimal strategy is to target the elderly. However, with a larger supply, if vaccination can occur while other interventions are maintained, the optimal strategy switches to targeting key transmitters to indirectly protect the vulnerable. As supply increases, vaccines that reduce or block infection have a greater impact than those that prevent disease alone due to the indirect protection provided to high-risk groups. Given a 2 billion global dose supply in 2021, we find that a strategy in which doses are allocated to countries proportional to population size is close to optimal in averting deaths and aligns with the ethical principles agreed in pandemic preparedness planning.