Bayesian phylogeography of influenza A/H3N2 for the 2014-15 season in the United States using three frameworks of ancestral state reconstruction

Commuting-driven competition between transmission chains shapes seasonal influenza virus epidemics in the United States

Despite intensive study, much remains unknown about the dynamics of seasonal influenza virus epidemic establishment and spread in the United States (US) each season. By reconstructing transmission lineages from seasonal influenza virus genomes collected in the US from 2014 to 2023, we show that most epidemics consisted of multiple distinct transmission lineages. Spread of these lineages exhibited strong spatiotemporal hierarchies and lineage size was correlated with timing of lineage establishment in the US. Mechanistic epidemic simulations suggest that mobility-driven competition between lineages determined the extent of individual lineages' geographical spread. Based on phylogeographic analyses and epidemic simulations, lineage-specific movement patterns were dominated by human commuting behavior. These results suggest that given the locations of early-season epidemic sparks, the topology of inter-state human mobility yields repeatable patterns of which influenza viruses will circulate where, but the importance of short-term processes limits predictability of regional and national epidemics.

Ecological and evolutionary dynamics of multi-strain RNA viruses

Potential interactions among co-circulating viral strains in host populations are often overlooked in the study of virus transmission. However, these interactions probably shape transmission dynamics by influencing host immune responses or altering the relative fitness among co-circulating strains. In this Review, we describe multi-strain dynamics from ecological and evolutionary perspectives, outline scales in which multi-strain dynamics occur and summarize important immunological, phylogenetic and mathematical modelling approaches used to quantify interactions among strains. We also discuss how host-pathogen interactions influence the co-circulation of pathogens. Finally, we highlight outstanding questions and knowledge gaps in the current theory and study of ecological and evolutionary dynamics of multi-strain viruses.

The impact of sampling bias on viral phylogeographic reconstruction

Genomic epidemiology plays an ever-increasing role in our understanding of and response to the spread of infectious pathogens. Phylogeography, the reconstruction of the historical location and movement of pathogens from the evolutionary relationships among sampled pathogen sequences, can inform policy decisions related to viral movement among jurisdictions. However, phylogeographic reconstruction is impacted by the fact that the sampling and virus sequencing policies differ among jurisdictions, and these differences can cause bias in phylogeographic reconstructions. Here we assess the potential impacts of geographic-based sampling bias on estimated viral locations in the past, and on whether key viral movements can be detected. We quantify the effect of bias using simulated phylogenies with known geographic histories, and determine the impact of the biased sampling and of the underlying migration rate on the accuracy of estimated past viral locations. We find that overall, the accuracy of phylogeographic reconstruction is high, particularly when the migration rate is low. However, results depend on sampling, and sampling bias can have a large impact on the numbers and nature of estimated migration events. We apply these insights to the geographic spread of Ebolavirus in the 2014-2016 West Africa epidemic. This work highlights how sampling policy can both impact geographic inference and be optimized to best ensure the accuracy of specific features of geographic spread.

Integrating animal movements with phylogeography to model the spread of PRRSV in the USA

Viral sequence data coupled with phylodynamic models have become instrumental in investigating the outbreaks of human and animal diseases, and the incorporation of the hypothesized drivers of pathogen spread can enhance the interpretation from phylodynamic inference. Integrating animal movement data with phylodynamics allows us to quantify the extent to which the spatial diffusion of a pathogen is influenced by animal movements and contrast the relative importance of different types of movements in shaping pathogen distribution. We combine animal movement, spatial, and environmental data in a Bayesian phylodynamic framework to explain the spatial diffusion and evolutionary trends of a rapidly spreading sub-lineage (denoted L1A) of porcine reproductive and respiratory syndrome virus (PRRSV) Type 2 from 2014 to 2017. PRRSV is the most important endemic pathogen affecting pigs in the USA, and this particular virulent sub-lineage emerged in 2014 and continues to be the dominant lineage in the US swine industry to date. Data included 984 open reading frame 5 (ORF5) PRRSV L1A sequences obtained from two production systems in a swine-dense production region (∼85,000 mi²) in the USA between 2014 and 2017. The study area was divided into sectors for which model covariates were summarized, and animal movement data between each sector were summarized by age class (wean: 3–4 weeks; feeder: 8–25 weeks; breeding: ≥21 weeks). We implemented a discrete-space phylogeographic generalized linear model using Bayesian evolutionary analysis by sampling trees (BEAST) to infer factors associated with variability in between-sector diffusion rates of PRRSV L1A. We found that between-sector spread was enhanced by the movement of feeder pigs, spatial adjacency of sectors, and farm density in the destination sector. The PRRSV L1A strain was introduced in the study area in early 2013, and genetic diversity and effective population size peaked in 2015 before fluctuating seasonally (peaking during the summer months). Our study underscores the importance of animal movements and shows, for the first time, that the movement of feeder pigs (8–25 weeks old) shaped the spatial patterns of PRRSV spread much more strongly than the movements of other age classes of pigs. The inclusion of movement data into phylodynamic models as done in this analysis may enhance our ability to identify crucial pathways of disease spread that can be targeted to mitigate the spatial spread of infectious human and animal pathogens.

Phylogeographic Approaches to Characterize the Emergence of Plant Pathogens

Phylogeography combines geographic information with phylogenetic and population genomic approaches to infer the evolutionary history of a species or population in a geographic context. This approach has been instrumental in understanding the emergence, spread, and evolution of a range of plant pathogens. In particular, phylogeography can address questions about where a pathogen originated, whether it is native or introduced, and when and how often introductions occurred. We review the theory, methods, and approaches underpinning phylogeographic inference and highlight applications providing novel insights into the emergence and spread of select pathogens. We hope that this review will be useful in assessing the power, pitfalls, and opportunities presented by various phylogeographic approaches.

The evolutionary history and global spatio-temporal dynamics of potato virus Y

Potato virus Y (PVY) is a destructive plant pathogen that causes considerable losses to global potato and tobacco production. Although the molecular structure of PVY is well characterized, the evolutionary and global transmission dynamics of this virus remain poorly understood. We investigated the phylodynamics of the virus by analysing 253 nucleotide sequences of the genes encoding the third protein (P3), cylindrical inclusion protein (CI), and the nuclear inclusion protein (NIb). Our Bayesian phylogenetic analyses showed that the mean substitution rates of different regions of the genome ranged from 8.50 × 10^-5 to 1.34 × 10^-4 substitutions/site/year, whereas the time to the most recent common ancestor of PVY varied with the length of the genomic regions and with the number of viral isolates being analysed. Our phylogeographic analysis showed that the PVY population originated in South America and was introduced into Europe in the 19th century, from where it spread around the globe. The migration pathways of PVY correlate well with the trade routes of potato tubers, suggesting that the global spread of PVY is associated with human activities.

Going back to the roots: Evaluating Bayesian phylogeographic models with discrete trait uncertainty

Phylogeography is a popular way to analyze virus sequences annotated with discrete, epidemiologically-relevant, trait data. For applied public health surveillance, a key quantity of interest is often the state at the root of the inferred phylogeny. In epidemiological terms, this represents the geographic origin of the observed outbreak. Since determining the origin of an outbreak is often critical for public health intervention, it is prudent to understand how well phylogeographic models perform this root state classification task under various analytical scenarios. Specifically, we investigate how discrete state space and sequence data set influence the root state classification accuracy. We performed phylogeographic inference on several simulated DNA data sets while i) increasing the number of sequences and ii) increasing the total number of possible discrete trait values. We show that phylogeographic models tend to perform best at intermediate sequence data set sizes. Further, we demonstrate that a popular metric used for evaluation of phylogeographic models, the Kullback-Leibler (KL) divergence, both increases with discrete state space and data set sizes. Further, by modeling phylogeographic root state classification accuracy using logistic regression, we show that KL is not supported as a predictor of model accuracy, indicating its limited utility for assessing phylogeographic model performance on empirical data. These results suggest that relying solely on the KL metric may lead to artificially inflated support for models with finer discretization schemes and larger data set sizes. These results will be important for public health practitioners seeking to use phylogeographic models for applied infectious disease surveillance.

Insights into Genomic Epidemiology, Evolution, and Transmission Dynamics of Genotype VII of Class II Newcastle Disease Virus in China

Newcastle disease virus (NDV) is distributed worldwide and has caused significant losses to the poultry industry. Almost all virulent NDV strains belong to class II, among which genotype VII is the predominant genotype in China. However, the molecular evolution and phylodynamics of class II genotype VII NDV strains in China remained largely unknown. In this study, we identified 13 virulent NDV including 11 genotype VII strains and 2 genotype IX strains, from clinical samples during 1997 to 2019. Combined NDV sequences submitted to GenBank, we investigate evolution, and transmission dynamics of class II NDVs in China, especially genotype VII strains. Our results revealed that East and South China have the most genotypic diversity of class II NDV, and East China might be the origin of genotype VII NDVs in China. In addition, genotype VII NDVs in China are presumably transmitted by chickens, as the virus was most prevalent in chickens. Furthermore, codon usage analysis revealed that the F genes of genotype VII NDVs have stronger adaptation in chickens, and six amino acids in this gene are found under positive selection via selection model analysis. Collectively, our results revealed the genetic diversity and evolutionary dynamics of genotype VII NDVs in China, providing important insights into the epidemiology of these viruses in China.

Evolutionary dynamics and transmission patterns of Newcastle disease virus in China through Bayesian phylogeographical analysis

The Chinese poultry industry has experienced outbreaks of Newcastle disease (ND) dating back to the 1920s. However, the epidemic has exhibited a downtrend in recent years. In this study, both observational and genetic data [fusion (F) and haemagglutinin-neuraminidase genes (HN)] were analyzed, and phylogeographic analysis based on prevalent genotypes of Newcastle disease virus (NDV) was conducted for better understanding of the evolution and spatiotemporal dynamics of ND in China. In line with the observed trend of epidemic outbreaks, the effective population size of F and HN genes of circulating NDV is no longer growing since 2000, which is supported by 95% highest posterior diversity (HPD) intervals. Phylogeographic analysis indicated that the two eastern coastal provinces, Shandong and Jiangsu were the most relevant hubs for NDV migration, and the geographical regions with active NDV diffusion seemed to be constrained to southern and eastern China. The live poultry trade may play an important role in viral spread. Interestingly, no migration links from wild birds to poultry received Bayes factor support (BF > 3), while the migration links from poultry to wild birds accounted for 64% in all effective migrations. This may indicate that the sporadic cases of ND in wild bird likely spillover events from poultry. These findings contribute to predictive models of NDV transmission, and potentially help in the prevention of future outbreaks.

Phylodynamic analyses of class I Newcastle disease virus isolated in China

Newcastle disease virus (NDV), the pathogen of Newcastle disease, has caused significant losses to the poultry industry worldwide. However, owing to its avirulence, class I NDVs have not been studied as much as class II NDVs. We aimed to epidemiologically monitor the spread of class I NDVs in China. We isolated 104 class I NDV strains from poultry in live poultry markets (LPMs) of Guangdong Province, south China, between January 2016 and December 2018. Genetic analysis revealed that all 104 isolates and most of the strains isolated from China were clustered into genotype 1.1.2 of class I NDVs. Bayesian analysis revealed that, although the United States may be the source, east and south China may be the epicentres of class I NDVs in China. In addition, in China, class I NDVs are presumably transmitted by chickens and domestic ducks as the virus is mostly prevalent in these birds. These novel findings demonstrated that class I NDVs are prevalent in south China, and it is important to perform routine surveillance and limit the numbers of different birds in different areas of LPMs to decrease the risk of intra- and interspecies transmission of NDVs.

Animal Disease Surveillance in the 21st Century: Applications and Robustness of Phylodynamic Methods in Recent U.S. Human-Like H3 Swine Influenza Outbreaks

Emerging and endemic animal viral diseases continue to impose substantial impacts on animal and human health. Most current and past molecular surveillance studies of animal diseases investigated spatio-temporal and evolutionary dynamics of the viruses in a disjointed analytical framework, ignoring many uncertainties and made joint conclusions from both analytical approaches. Phylodynamic methods offer a uniquely integrated platform capable of inferring complex epidemiological and evolutionary processes from the phylogeny of viruses in populations using a single Bayesian statistical framework. In this study, we reviewed and outlined basic concepts and aspects of phylodynamic methods and attempted to summarize essential components of the methodology in one analytical pipeline to facilitate the proper use of the methods by animal health researchers. Also, we challenged the robustness of the posterior evolutionary parameters, inferred by the commonly used phylodynamic models, using hemagglutinin (HA) and polymerase basic 2 (PB2) segments of the currently circulating human-like H3 swine influenza (SI) viruses isolated in the United States and multiple priors. Subsequently, we compared similarities and differences between the posterior parameters inferred from sequence data using multiple phylodynamic models. Our suggested phylodynamic approach attempts to reduce the impact of its inherent limitations to offer less biased and biologically plausible inferences about the pathogen evolutionary characteristics to properly guide intervention activities. We also pinpointed requirements and challenges for integrating phylodynamic methods in routine animal disease surveillance activities.

Domestic ducks play a major role in the maintenance and spread of H5N8 highly pathogenic avian influenza viruses in South Korea

The H5N8 highly pathogenic avian influenza viruses (HPAIVs) belonging to clade 2.3.4.4 spread from Eastern China to Korea in 2014 and caused outbreaks in domestic poultry until 2016. To understand how H5N8 HPAIVs spread at host species level in Korea during 2014-2016, a Bayesian phylogenetic analysis was used for ancestral state reconstruction and estimation of the host transition dynamics between wild waterfowl, domestic ducks and chickens. Our data support that H5N8 HPAIV most likely transmitted from wild waterfowl to domestic ducks, and then maintained in domestic ducks followed by dispersal of HPAIV from domestic ducks to chickens, suggesting domestic duck population plays a central role in the maintenance, amplification and spread of wild HPAIV to terrestrial poultry in Korea.

A Fast Likelihood Method to Reconstruct and Visualize Ancestral Scenarios

The reconstruction of ancestral scenarios is widely used to study the evolution of characters along phylogenetic trees. One commonly uses the marginal posterior probabilities of the character states, or the joint reconstruction of the most likely scenario. However, marginal reconstructions provide users with state probabilities, which are difficult to interpret and visualize, whereas joint reconstructions select a unique state for every tree node and thus do not reflect the uncertainty of inferences. We propose a simple and fast approach, which is in between these two extremes. We use decision-theory concepts (namely, the Brier score) to associate each node in the tree to a set of likely states. A unique state is predicted in tree regions with low uncertainty, whereas several states are predicted in uncertain regions, typically around the tree root. To visualize the results, we cluster the neighboring nodes associated with the same states and use graph visualization tools. The method is implemented in the PastML program and web server. The results on simulated data demonstrate the accuracy and robustness of the approach. PastML was applied to the phylogeography of Dengue serotype 2 (DENV2), and the evolution of drug resistances in a large HIV data set. These analyses took a few minutes and provided convincing results. PastML retrieved the main transmission routes of human DENV2 and showed the uncertainty of the human-sylvatic DENV2 geographic origin. With HIV, the results show that resistance mutations mostly emerge independently under treatment pressure, but resistance clusters are found, corresponding to transmissions among untreated patients.

Phylodynamics of Influenza A/H1N1pdm09 in India Reveals Circulation Patterns and Increased Selection for Clade 6b Residues and Other High Mortality Mutants

The clinical severity and observed case fatality ratio of influenza A/H1N1pdm09 in India, particularly in 2015 and 2017 far exceeds current global estimates. Reasons for these frequent and severe epidemic waves remain unclear. We used Bayesian phylodynamic methods to uncover possible genetic explanations for this, while also identifying the transmission dynamics of A/H1N1pdm09 between 2009 and 2017 to inform future public health interventions. We reveal a disproportionate selection at haemagglutinin residue positions associated with increased morbidity and mortality in India such as position 222 and clade 6B characteristic residues, relative to equivalent isolates circulating globally. We also identify for the first time, increased selection at position 186 as potentially explaining the severity of recent A/H1N1pdm09 epidemics in India. We reveal national routes of A/H1N1pdm09 transmission, identifying Maharashtra as the most important state for the spread throughout India, while quantifying climactic, ecological, and transport factors as drivers of within-country transmission. Together these results have important implications for future A/H1N1pdm09 surveillance and control within India, but also for epidemic and pandemic risk prediction around the world.

Bayesian phylodynamics of avian influenza A virus H9N2 in Asia with time-dependent predictors of migration

Model-based phylodynamic approaches recently employed generalized linear models (GLMs) to uncover potential predictors of viral spread. Very recently some of these models have allowed both the predictors and their coefficients to be time-dependent. However, these studies mainly focused on predictors that are assumed to be constant through time. Here we inferred the phylodynamics of avian influenza A virus H9N2 isolated in 12 Asian countries and regions under both discrete trait analysis (DTA) and structured coalescent (MASCOT) approaches. Using MASCOT we applied a new time-dependent GLM to uncover the underlying factors behind H9N2 spread. We curated a rich set of time-series predictors including annual international live poultry trade and national poultry production figures. This time-dependent phylodynamic prediction model was compared to commonly employed time-independent alternatives. Additionally the time-dependent MASCOT model allowed for the estimation of viral effective sub-population sizes and their changes through time, and these effective population dynamics within each country were predicted by a GLM. International annual poultry trade is a strongly supported predictor of virus migration rates. There was also strong support for geographic proximity as a predictor of migration rate in all GLMs investigated. In time-dependent MASCOT models, national poultry production was also identified as a predictor of virus genetic diversity through time and this signal was obvious in mainland China. Our application of a recently introduced time-dependent GLM predictors integrated rich time-series data in Bayesian phylodynamic prediction. We demonstrated the contribution of poultry trade and geographic proximity (potentially unheralded wild bird movements) to avian influenza spread in Asia. To gain a better understanding of the drivers of H9N2 spread, we suggest increased surveillance of the H9N2 virus in countries that are currently under-sampled as well as in wild bird populations in the most affected countries.

Phylogeographical and cross-species transmission dynamics of SAT1 and SAT2 foot-and-mouth disease virus in Eastern Africa

Understanding the dynamics of foot-and-mouth disease virus (FMDV), an endemic and economically constraining disease, is critical in designing control programmes in Africa. This study investigates the evolutionary epidemiology of SAT1 and SAT2 FMDV in Eastern Africa, as well as between cattle and wild African buffalo. Bayesian phylodynamic models were used to analyse SAT1 and SAT2 VP1 gene segments collected between 1975 and 2016, focusing on the SAT1 and SAT2 viruses currently circulating in Eastern Africa. The root state posterior probabilities inferred from our analyses suggest Zimbabwe as the ancestral location for SAT1 currently circulating in Eastern Africa (p = 0.67). For the SAT2 clade, Kenya is inferred to be the ancestral location for introduction of the virus into other countries in Eastern Africa (p = 0.72). Salient (Bayes factor >10) viral dispersal routes were inferred from Tanzania to Kenya, and from Kenya to Uganda for SAT1 and SAT2, respectively. Results suggest that cattle are the source of the SAT1 and SAT2 clades currently circulating in Eastern Africa. In addition, our results suggest that the majority of SAT1 and SAT2 in livestock come from other livestock rather than wildlife, with limited evidence that buffalo serve as reservoirs for cattle. Insights from the present study highlight the role of cattle movements and anthropogenic activities in shaping the evolutionary history of SAT1 and SAT2 in Eastern Africa. While the results may be affected by inherent limitations of imperfect surveillance, our analysis elucidates the dynamics between host species in this region, which is key to guiding disease intervention activities.

Characterising routes of H5N1 and H7N9 spread in China using Bayesian phylogeographical analysis

Avian influenza H5N1 subtype has caused a global public health concern due to its high pathogenicity in poultry and high case fatality rates in humans. The recently emerged H7N9 is a growing pandemic risk due to its sustained high rates of human infections, and recently acquired high pathogenicity in poultry. Here, we used Bayesian phylogeography on 265 H5N1 and 371 H7N9 haemagglutinin sequences isolated from humans, animals and the environment, to identify and compare migration patterns and factors predictive of H5N1 and H7N9 diffusion rates in China. H7N9 diffusion dynamics and predictor contributions differ from H5N1. Key determinants of spatial diffusion included: proximity between locations (for H5N1 and H7N9), and lower rural population densities (H5N1 only). For H7N9, additional predictors included low avian influenza vaccination rates, low percentage of nature reserves and high humidity levels. For both H5N1 and H7N9, we found viral migration rates from Guangdong to Guangxi and Guangdong to Hunan were highly supported transmission routes (Bayes Factor > 30). We show fundamental differences in wide-scale transmission dynamics between H5N1 and H7N9. Importantly, this indicates that avian influenza initiatives designed to control H5N1 may not be sufficient for controlling the H7N9 epidemic. We suggest control and prevention activities to specifically target poultry transportation networks between Central, Pan-Pearl River Delta and South-West regions.

On the importance of negative controls in viral landscape phylogeography

Phylogeographic reconstructions are becoming an established procedure to evaluate the factors that could impact virus spread. While a discrete phylogeographic approach can be used to test predictors of transition rates among discrete locations, alternative continuous phylogeographic reconstructions can also be exploited to investigate the impact of underlying environmental layers on the dispersal velocity of a virus. The two approaches are complementary tools for studying pathogens' spread, but in both cases, care must be taken to avoid misinterpretations. Here, we analyse rice yellow mottle virus (RYMV) sequence data from West and East Africa to illustrate how both approaches can be used to study the impact of environmental factors on the virus’ dispersal frequency and velocity. While it was previously reported that host connectivity was a major determinant of RYMV spread, we show that this was a false positive result due to the lack of appropriate negative controls. We also discuss and compare the phylodynamic tools currently available for investigating the impact of environmental factors on virus spread.

The effects of random taxa sampling schemes in Bayesian virus phylogeography

Public health researchers are often tasked with accurately and quickly identifying the location and time when an epidemic originated from a representative sample of nucleotide sequences. In this paper, we investigate multiple approaches to subsampling the sequence set when employing a Bayesian phylogeographic generalized linear model. Our results indicate that near-categorical posterior MCC estimates on the root can be obtained with replicate runs using 25-50% of the sequence data, and that including 90% of sequences does not necessarily entail more accurate inferences. We present the first analysis of predictor signal suppression and show how the ability to detect the influence of predictor variables is limited when sample size predictors are included in the models.

Phylogeography of H5N1 avian influenza virus in Indonesia

Highly pathogenic avian influenza (HPAI) viruses of the H5N1 subtype are a major concern to human and animal health in Indonesia. This study aimed to characterize transmission dynamics of H5N1 over time using novel Bayesian phylogeography methods to identify factors which have influenced the spread of H5N1 in Indonesia. We used publicly available hemagglutinin sequence data sampled between 2003 and 2016 to model ancestral state reconstruction of HPAI H5N1 evolution. We found strong support for H5N1 transmission routes between provinces in Java Island and inter-island transmissions, such as between Nusa Tenggara and Kalimantan Islands, not previously described. The spread is consistent with wild bird flyways and poultry trading routes. H5N1 migration was associated with the regions of high chicken densities and low human development indices. These results can be used to inform more targeted planning of H5N1 control and prevention activities in Indonesia.

The Effects of Sampling Location and Predictor Point Estimate Certainty on Posterior Support in Bayesian Phylogeographic Generalized Linear Models

The use of generalized linear models in Bayesian phylogeography has enabled researchers to simultaneously reconstruct the spatiotemporal history of a virus and quantify the contribution of predictor variables to that process. However, little is known about the sensitivity of this method to the choice of the discrete state partition. Here we investigate this question by analyzing a data set containing 299 sequences of the West Nile virus envelope gene sampled in the United States and fifteen predictors aggregated at four spatial levels. We demonstrate that although the topology of the viral phylogenies was consistent across analyses, support for the predictors depended on the level of aggregation. In particular, we found that the variance of the predictor support metrics was minimized at the most precise level for several predictors and maximized at more sparse levels of aggregation. These results suggest that caution should be taken when partitioning a region into discrete locations to ensure that interpretable, reproducible posterior estimates are obtained. These results also demonstrate why researchers should use the most precise discrete states possible to minimize the posterior variance in such estimates and reveal what truly drives the diffusion of viruses.