Identifying selection in the within-host evolution of influenza using viral sequence data

Variable rates of SARS-CoV-2 evolution in chronic infections

An important feature of the evolution of the SARS-CoV-2 virus has been the emergence of highly mutated novel variants, which are characterised by the gain of multiple mutations relative to viruses circulating in the general global population. Cases of chronic viral infection have been suggested as an explanation for this phenomenon, whereby an extended period of infection, with an increased rate of evolution, creates viruses with substantial genetic novelty. However, measuring a rate of evolution during chronic infection is made more difficult by the potential existence of compartmentalisation in the viral population, whereby the viruses in a host form distinct subpopulations. We here describe and apply a novel statistical method to study within-host virus evolution, identifying the minimum number of subpopulations required to explain sequence data observed from cases of chronic infection, and inferring rates for within-host viral evolution. Across nine cases of chronic SARS-CoV-2 infection in hospitalised patients we find that non-trivial population structure is relatively common, with five cases showing evidence of more than one viral population evolving independently within the host. The detection of non-trivial population structure was more common in severely immunocompromised individuals (p = 0.04, Fisher's Exact Test). We find cases of within-host evolution proceeding significantly faster, and significantly slower, than that of the global SARS-CoV-2 population, and of cases in which viral subpopulations in the same host have statistically distinguishable rates of evolution. Non-trivial population structure was associated with high rates of within-host evolution that were systematically underestimated by a more standard inference method.

Quantifying the impact of vaccination on transmission and diversity of influenza A variants in pigs

Global evolutionary dynamics of influenza A virus (IAV) are fundamentally driven by the extent of virus diversity generated, transmitted, and shaped in individual hosts. How vaccination affects the degree of IAV genetic diversity that can be transmitted and expanded in pigs is unknown. To evaluate the effect of vaccination on the transmission of genetically distinct IAV variants and their diversity after transmission in pigs, we examined the whole genome of IAV recovered from the nasal cavities of pigs vaccinated with different influenza immunization regimens after being infected simultaneously by H1N1 and H3N2 IAVs using a seeder pig model. We found that the seeder pigs harbored more diversified virus populations than the contact pigs. Among contact pigs, H3N2 and H1N1 viruses recovered from pigs vaccinated with a single dose of an unmatched modified live vaccine generally accumulated more extensive genetic mutations than non-vaccinated pigs. Furthermore, the non-sterilizing immunity elicited by the single-dose-modified live vaccine may have exerted positive selection on H1 antigenic regions as we detected significantly higher nonsynonymous but lower synonymous evolutionary rates in H1 antigenic regions than non-antigenic regions. In addition, we observed that the vaccinated pigs shared significantly less proportion of H3N2 variants with seeder pigs than unvaccinated pigs. These results indicated that vaccination might reduce the impact of transmitted influenza variants on the overall diversity of IAV populations harbored in recipient pigs and that within-host genetic selection of IAV is more likely to occur in pigs vaccinated with improperly matched vaccines.IMPORTANCEUnderstanding how vaccination shapes the diversity of influenza variants that transmit and propagate among pigs is essential for designing effective IAV surveillance and control programs. Current knowledge about the transmission of IAV variants has primarily been explored in humans during natural infection. However, how immunity elicited by improperly matched vaccines affects the degree of IAV genetic diversity that can be transmitted and expanded in pigs at the whole-genome level is unknown. We analyzed IAV sequences from samples collected daily from experimentally infected pigs vaccinated with various protocols in a field-represented IAV co-infection model. We found that vaccine-induced non-sterilizing immunity might promote genetic variation on the IAV genome and drive positive selection at antigenic sites during infection. In addition, a smaller proportion of H3N2 viral variants were shared between seeder pigs and vaccinated pigs, suggesting the influence of vaccination on shaping the virus genomic diversity in recipient pigs during the transmission events.

Fast and Accurate Estimation of Selection Coefficients and Allele Histories from Ancient and Modern DNA

We here present CLUES2, a full-likelihood method to infer natural selection from sequence data that is an extension of the method CLUES. We make several substantial improvements to the CLUES method that greatly increases both its applicability and its speed. We add the ability to use ancestral recombination graphs on ancient data as emissions to the underlying hidden Markov model, which enables CLUES2 to use both temporal and linkage information to make estimates of selection coefficients. We also fully implement the ability to estimate distinct selection coefficients in different epochs, which allows for the analysis of changes in selective pressures through time, as well as selection with dominance. In addition, we greatly increase the computational efficiency of CLUES2 over CLUES using several approximations to the forward-backward algorithms and develop a new way to reconstruct historic allele frequencies by integrating over the uncertainty in the estimation of the selection coefficients. We illustrate the accuracy of CLUES2 through extensive simulations and validate the importance sampling framework for integrating over the uncertainty in the inference of gene trees. We also show that CLUES2 is well-calibrated by showing that under the null hypothesis, the distribution of log-likelihood ratios follows a χ2 distribution with the appropriate degrees of freedom. We run CLUES2 on a set of recently published ancient human data from Western Eurasia and test for evidence of changing selection coefficients through time. We find significant evidence of changing selective pressures in several genes correlated with the introduction of agriculture to Europe and the ensuing dietary and demographic shifts of that time. In particular, our analysis supports previous hypotheses of strong selection on lactase persistence during periods of ancient famines and attenuated selection in more modern periods.

Correlated Allele Frequency Changes Reveal Clonal Structure and Selection in Temporal Genetic Data

In evolving populations where the rate of beneficial mutations is large, subpopulations of individuals with competing beneficial mutations can be maintained over long times. Evolution with this kind of clonal structure is commonly observed in a wide range of microbial and viral populations. However, it can be difficult to completely resolve clonal dynamics in data. This is due to limited read lengths in high-throughput sequencing methods, which are often insufficient to directly measure linkage disequilibrium or determine clonal structure. Here, we develop a method to infer clonal structure using correlated allele frequency changes in time-series sequence data. Simulations show that our method recovers true, underlying clonal structures when they are known and accurately estimate linkage disequilibrium. This information can then be combined with other inference methods to improve estimates of the fitness effects of individual mutations. Applications to data suggest novel clonal structures in an E. coli long-term evolution experiment, and yield improved predictions of the effects of mutations on bacterial fitness and antibiotic resistance. Moreover, our method is computationally efficient, requiring orders of magnitude less run time for large data sets than existing methods. Overall, our method provides a powerful tool to infer clonal structures from data sets where only allele frequencies are available, which can also improve downstream analyses.

Antigenic cooperation in viral populations: Transformation of functions of intra-host viral variants

In this paper, we study intra-host viral adaptation by antigenic cooperation - a mechanism of immune escape that serves as an alternative to the standard mechanism of escape by continuous genomic diversification and allows to explain a number of experimental observations associated with the establishment of chronic infections by highly mutable viruses. Within this mechanism, the topology of a cross-immunoreactivity network forces intra-host viral variants to specialize for complementary roles and adapt to the host's immune response as a quasi-social ecosystem. Here we study dynamical changes in immune adaptation caused by evolutionary and epidemiological events. First, we show that the emergence of a viral variant with altered antigenic features may result in a rapid re-arrangement of the viral ecosystem and a change in the roles played by existing viral variants. In particular, it may push the population under immune escape by genomic diversification towards the stable state of adaptation by antigenic cooperation. Next, we study the effect of a viral transmission between two chronically infected hosts, which results in the merging of two intra-host viral populations in the state of stable immune-adapted equilibrium. In this case, we also describe how the newly formed viral population adapts to the host's environment by changing the functions of its members. The results are obtained analytically for minimal cross-immunoreactivity networks and numerically for larger populations.

Local Immunodeficiency: Combining Cross-Immunoreactivity Networks

This article continues the analysis of the recently observed phenomenon of local immunodeficiency (LI), which arises as a result of antigenic cooperation among intrahost viruses organized into a network of cross-immunoreactivity (CR). We study here what happens as the result of combining (connecting) the simplest CR networks, which have a stable state of LI. It turned out that many possibilities occur, particularly resulting in a change of roles of some viruses in the CR network. Our results also give some indications about a boundary of the set of CR networks with stable state of LI in the entire collection of all possible CR networks. Such borderline CR networks are characterized by only a marginally stable (neutral rather than stable) state of the LI, or by the existence of such subnetworks in a CR network that evolve independently of each other (although being connected).

Viral intra-host evolutionary dynamics revealed via serial passage of Japanese encephalitis virus in vitro

Analyses of viral inter- and intra-host mutations could better guide the prevention and control of infectious diseases. For a long time, studies on viral evolution have focused on viral inter-host variations. Next-generation sequencing has accelerated the investigations of viral intra-host diversity. However, the theoretical basis and dynamic characteristics of viral intra-host mutations remain unknown. Here, using serial passages of the SA14-14-2 vaccine strain of Japanese encephalitis virus (JEV) as the in vitro model, the distribution characteristics of 1,788 detected intra-host single-nucleotide variations (iSNVs) and their mutated frequencies from 477 deep-sequenced samples were analyzed. Our results revealed that in adaptive (baby hamster kidney (BHK)) cells, JEV is under a nearly neutral selection pressure, and both non-synonymous and synonymous mutations represent an S-shaped growth trend over time. A higher positive selection pressure was observed in the nonadaptive (C6/36) cells, and logarithmic growth in non-synonymous iSNVs and linear growth in synonymous iSNVs were observed over time. Moreover, the mutation rates of the NS4B protein and the untranslated region (UTR) of the JEV are significantly different between BHK and C6/36 cells, suggesting that viral selection pressure is regulated by different cellular environments. In addition, no significant difference was detected in the distribution of mutated frequencies of iSNVs between BHK and C6/36 cells.

Estimating linkage disequilibrium and selection from allele frequency trajectories

Genetic sequences collected over time provide an exciting opportunity to study natural selection. In such studies, it is important to account for linkage disequilibrium to accurately measure selection and to distinguish between selection and other effects that can cause changes in allele frequencies, such as genetic hitchhiking or clonal interference. However, most high-throughput sequencing methods cannot directly measure linkage due to short-read lengths. Here we develop a simple method to estimate linkage disequilibrium from time-series allele frequencies. This reconstructed linkage information can then be combined with other inference methods to infer the fitness effects of individual mutations. Simulations show that our approach reliably outperforms inference that ignores linkage disequilibrium and, with sufficient sampling, performs similarly to inference using the true linkage information. We also introduce two regularization methods derived from random matrix theory that help to preserve its performance under limited sampling effects. Overall, our method enables the use of linkage-aware inference methods even for data sets where only allele frequency time series are available.

Inferring Epistasis from Genetic Time-series Data

Epistasis refers to fitness or functional effects of mutations that depend on the sequence background in which these mutations arise. Epistasis is prevalent in nature, including populations of viruses, bacteria, and cancers, and can contribute to the evolution of drug resistance and immune escape. However, it is difficult to directly estimate epistatic effects from sampled observations of a population. At present, there are very few methods that can disentangle the effects of selection (including epistasis), mutation, recombination, genetic drift, and genetic linkage in evolving populations. Here we develop a method to infer epistasis, along with the fitness effects of individual mutations, from observed evolutionary histories. Simulations show that we can accurately infer pairwise epistatic interactions provided that there is sufficient genetic diversity in the data. Our method also allows us to identify which fitness parameters can be reliably inferred from a particular data set and which ones are unidentifiable. Our approach therefore allows for the inference of more complex models of selection from time-series genetic data, while also quantifying uncertainty in the inferred parameters.

Vaccination decreases the risk of influenza A virus reassortment but not genetic variation in pigs

Although vaccination is broadly used in North American swine breeding herds, managing swine influenza is challenging primarily due to the continuous evolution of influenza A virus (IAV) and the ability of the virus to transmit among vaccinated pigs. Studies that have simultaneously assessed the impact of vaccination on the emergence of IAV reassortment and genetic variation in pigs are limited. Here, we directly sequenced 28 bronchoalveolar lavage fluid (BALF) samples collected from vaccinated and unvaccinated pigs co-infected with H1N1 and H3N2 IAV strains, and characterized 202 individual viral plaques recovered from 13 BALF samples. We identified 54 reassortant viruses that were grouped in 17 single and 16 mixed genotypes. Notably, we found that prime-boost vaccinated pigs had less reassortant viruses than nonvaccinated pigs, likely due to a reduction in the number of days pigs were co-infected with both challenge viruses. However, direct sequencing from BALF samples revealed limited impact of vaccination on viral variant frequency, evolutionary rates, and nucleotide diversity in any IAV coding regions. Overall, our results highlight the value of IAV vaccination not only at limiting virus replication in pigs but also at protecting public health by restricting the generation of novel reassortants with zoonotic and/or pandemic potential.

Swine influenza A viruses cause severe illness among pigs and financial losses on pig farms worldwide. These viruses can also infect humans and have caused deadly human pandemics in the past. Influenza A viruses are dangerous because viruses can be transferred between humans, birds and pigs. These co-infections can allow the viruses to swap genetic material. Viral genetic exchanges can result in new virus strains that are more dangerous or that can infect other types of animals more easily.

Farmers vaccinate their pigs to control the swine influenza A virus. The vaccines are regularly updated to match circulating virus strains. But the virus evolves rapidly to escape vaccine-induced immunity, and infections are common even in vaccinated pigs. Learning about how vaccination affects the evolution of influenza A viruses in pigs could help scientists prevent outbreaks on pig farms and avoid spillover pandemics in humans.

Li et al. show that influenza A viruses are less likely to swap genetic material in vaccinated and boosted pigs than in unvaccinated animals. In the experiments, Li et al. collected swine influenza A samples from the lungs of pigs that had received different vaccination protocols. Next, Li et al. used next-generation sequencing to identify new mutations in the virus or genetic swaps among different strains. In pigs infected with both the H1N1 and H3N2 strains of influenza, the two viruses began trading genes within a week. But less genetic mixing occurred in vaccinated and boosted pigs because they spent less time infected with both viruses than in unvaccinated pigs. The vaccination status of the pig did not have much effect on how many new mutations occurred in the viruses.

The experiments show that vaccinating and boosting pigs against influenza A viruses may protect against genetic swapping among influenza viruses. If future studies on pig farms confirm the results, the information gleaned from the study could help scientists improve farm vaccine protocols to further reduce influenza risks to animals and people.

Clinical sequencing uncovers the genomic characteristics and mutation spectrum of the 2018 African swine fever virus in Guangdong, China

African swine fever (ASF) outbreak have caused tremendous economic loss to the pig industry in China since its emergence in August 2018. Previous studies revealed that many published sequences are not suitable for detailed analyses due to the lack of data regarding quality parameters and methodology, and outdated annotations. Thus, high-quality genomes of highly pathogenic strains that can be used as references for early Chinese ASF outbreaks are still lacking, and little is known about the features of intra-host variants of ASF virus (ASFV). In this study, a full genome sequencing of clinical samples from the first ASF outbreak in Guangdong in 2018 was performed using MGI (MGI Tech Co., Ltd., Shenzhen, China) and Nanopore sequencing platforms, followed by Sanger sequencing to verify the variations. With 22 sequencing corrections, we obtained a high-quality genome of one of the earliest virulent isolates, GZ201801_2. After proofreading, we improved (add or modify) the annotations of this isolate using the whole genome alignment with Georgia 2007/1. Based on the complete genome sequence, we constructed the methylation profiles of early ASFV strains in China and predicted the potential 5mC and 6mA methylation sites, which are likely involved in metabolism, transcription, and replication. Additionally, the intra-host single nucleotide variant distribution and mutant allele frequency in the clinical samples of early strain were determined for the first time and found a strong preference for A and T substitution mutation, non-synonymous mutations, and mutations that resulted in amino acid substitutions into Lysine. In conclusion, this study provides a high-quality genome sequence, updated genome annotation, methylation profile, and mutation spectrum of early ASFV strains in China, thereby providing a reference basis for further studies on the evolution, transmission, and virulence of ASFV.

Two-step fitness selection for intra-host variations in SARS-CoV-2

Spontaneous mutations introduce uncertainty into coronavirus disease 2019 (COVID-19) control procedures and vaccine development. Here, we perform a spatiotemporal analysis on intra-host single-nucleotide variants (iSNVs) in 402 clinical samples from 170 affected individuals, which reveals an increase in genetic diversity over time after symptom onset in individuals. Nonsynonymous mutations are overrepresented in the pool of iSNVs but underrepresented at the single-nucleotide polymorphism (SNP) level, suggesting a two-step fitness selection process: a large number of nonsynonymous substitutions are generated in the host (positive selection), and these substitutions tend to be unfixed as SNPs in the population (negative selection). Dynamic iSNV changes in subpopulations with different gender, age, illness severity, and viral shedding time displayed a varied fitness selection process among populations. Our study highlights that iSNVs provide a mutational pool shaping the rapid global evolution of the virus.

MPL resolves genetic linkage in fitness inference from complex evolutionary histories

Genetic linkage causes the fate of new mutations in a population to be contingent on the genetic background on which they appear. This makes it challenging to identify how individual mutations affect fitness. To overcome this challenge, we developed marginal path likelihood (MPL), a method to infer selection from evolutionary histories that resolves genetic linkage. Validation on real and simulated data sets shows that MPL is fast and accurate, outperforming existing inference approaches. We found that resolving linkage is crucial for accurately quantifying selection in complex evolving populations, which we demonstrate through a quantitative analysis of intrahost HIV-1 evolution using multiple patient data sets. Linkage effects generated by variants that sweep rapidly through the population are particularly strong, extending far across the genome. Taken together, our results argue for the importance of resolving linkage in studies of natural selection.

Evolutionary dynamics of Tomato spotted wilt virus within and between alternate plant hosts and thrips

Tomato spotted wilt virus (TSWV) is a generalist pathogen with one of the broadest known host ranges among RNA viruses. To understand how TSWV adapts to different hosts, we experimentally passaged viral populations between two alternate hosts, Emilia sochifolia and Datura stramonium, and an obligate vector in which it also replicates, western flower thrips (Frankliniella occidentalis). Deep sequencing viral populations at multiple time points allowed us to track the evolutionary dynamics of viral populations within and between hosts. High levels of viral genetic diversity were maintained in both plants and thrips between transmission events. Rapid fluctuations in the frequency of amino acid variants indicated strong host-specific selection pressures on proteins involved in viral movement (NSm) and replication (RdRp). While several genetic variants showed opposing fitness effects in different hosts, fitness effects were generally positively correlated between hosts indicating that positive rather than antagonistic pleiotropy is pervasive. These results suggest that high levels of genetic diversity together with the positive pleiotropic effects of mutations have allowed TSWV to rapidly adapt to new hosts and expand its host range.

A large effective population size for established within-host influenza virus infection

Strains of the influenza virus form coherent global populations, yet exist at the level of single infections in individual hosts. The relationship between these scales is a critical topic for understanding viral evolution. Here we investigate the within-host relationship between selection and the stochastic effects of genetic drift, estimating an effective population size of infection Ne for influenza infection. Examining whole-genome sequence data describing a chronic case of influenza B in a severely immunocompromised child we infer an Ne of 2.5 × 107 (95% confidence range 1.0 × 107 to 9.0 × 107) suggesting that genetic drift is of minimal importance during an established influenza infection. Our result, supported by data from influenza A infection, suggests that positive selection during within-host infection is primarily limited by the typically short period of infection. Atypically long infections may have a disproportionate influence upon global patterns of viral evolution.

A de novo approach to inferring within-host fitness effects during untreated HIV-1 infection

In the absence of effective antiviral therapy, HIV-1 evolves in response to the within-host environment, of which the immune system is an important aspect. During the earliest stages of infection, this process of evolution is very rapid, driven by a small number of CTL escape mutations. As the infection progresses, immune escape variants evolve under reduced magnitudes of selection, while competition between an increasing number of polymorphic alleles (i.e., clonal interference) makes it difficult to quantify the magnitude of selection acting upon specific variant alleles. To tackle this complex problem, we developed a novel multi-locus inference method to evaluate the role of selection during the chronic stage of within-host infection. We applied this method to targeted sequence data from the p24 and gp41 regions of HIV-1 collected from 34 patients with long-term untreated HIV-1 infection. We identify a broad distribution of beneficial fitness effects during infection, with a small number of variants evolving under strong selection and very many variants evolving under weaker selection. The uniquely large number of infections analysed granted a previously unparalleled statistical power to identify loci at which selection could be inferred to act with statistical confidence. Our model makes no prior assumptions about the nature of alleles under selection, such that any synonymous or non-synonymous variant may be inferred to evolve under selection. However, the majority of variants inferred with confidence to be under selection were non-synonymous in nature, and in most cases were have previously been associated with either CTL escape in p24 or neutralising antibody escape in gp41. We also identified a putative new CTL escape site (residue 286 in gag), and a region of gp41 (including residues 644, 648, 655 in env) likely to be associated with immune escape. Sites inferred to be under selection in multiple hosts have high within-host and between-host diversity although not all sites with high between-host diversity were inferred to be under selection at the within-host level. Our identification of selection at sites associated with resistance to broadly neutralising antibodies (bNAbs) highlights the need to fully understand the role of selection in untreated individuals when designing bNAb based therapies.

Evolution of H9N2 low pathogenic avian influenza virus during passages in chickens

The process of avian influenza virus (AIV) evolution in a new host was investigated in the experiment in which ten serial passages of a turkey-derived H9N2 AIV were carried out in specific pathogen free chickens (3 birds/group) inoculated by oculonasal route. Oropharyngeal swabs collected 3 days post infection were used for inoculation of birds in the next passage and subjected to analysis using deep sequencing. In total, eight mutations in the consensus sequence were found in the viral pool derived from the 10th passage: four mutations (2 in PB1 and 2 in HA) were present in the inoculum as minority variants while the other four (2 in NP, 1 in PA and 1 in HA) emerged during the passages in chickens. The detected fluctuations in the genetic heterogeneity of viral pools from consecutive passages were most likely attributed to the selective bottleneck. The genes known for bearing molecular determinants of the AIV host specificity (HA, PB2, PB1, PA) contributed most to the overall virus diversity. In some cases, a fast selection of the novel variant was noticed. For example, the amino-acid substitution N337K in the haemagglutinin (HA) cleavage site region detected in the 6th passage as low frequency variant had undergone rapid selection and became predominant in the 7th passage. Interestingly, detection of identical mutation in the field H9N2 isolates 1-year apart suggests that this substitution might provide the virus with a selective advantage. However, the role of specific mutations and their influence on the virus adaptation or fitness are mostly unknown and require further investigations.

Immune responses induced by different genotypes of the disease-specific protein of Rice stripe virus in the vector insect

Persistent plant viruses circulate between host plants and vector insects, possibly leading to the genetic divergence in viral populations. We analyzed the single nucleotide polymorphisms (SNPs) of Rice stripe virus (RSV) when it incubated in the small brown planthopper and rice. Two SNPs, which lead to nonsynonymous substitutions in the disease-specific protein (SP) of RSV, produced three genotypes, i.e., GG, AA and GA. The GG type mainly existed in the early infection period of RSV in the planthoppers and was gradually substituted by the other two genotypes during viral transmission. The two SNPs did not affect the interactions of SP with rice PsbP or with RSV coat protein. The GG genotype of SP induced stronger immune responses than those of the other two genotypes in the pattern recognition molecule and immune-responsive effector pathways. These findings demonstrated the population variations of RSV during the circulation between the vector insect and host plant.

Mutational load causes stochastic evolutionary outcomes in acute RNA viral infection

Mutational load is known to be of importance for the evolution of RNA viruses, the combination of a high mutation rate and large population size leading to an accumulation of deleterious mutations. However, while the effects of mutational load on global viral populations have been considered, its quantitative effects at the within-host scale of infection are less well understood. We here show that even on the rapid timescale of acute disease, mutational load has an effect on within-host viral adaptation, reducing the effective selection acting upon beneficial variants by ∼10 per cent. Furthermore, mutational load induces considerable stochasticity in the pattern of evolution, causing a more than five-fold uncertainty in the effective fitness of a transmitted beneficial variant. Our work aims to bridge the gap between classic models from population genetic theory and the biology of viral infection. In an advance on some previous models of mutational load, we replace the assumption of a constant variant fitness cost with an experimentally-derived distribution of fitness effects. Expanding previous frameworks for evolutionary simulation, we introduce the Wright-Fisher model with continuous mutation, which describes a continuum of possible modes of replication within a cell. Our results advance our understanding of adaptation in the context of strong selection and a high mutation rate. Despite viral populations having large absolute sizes, critical events in viral adaptation, including antigenic drift and the onset of drug resistance, arise through stochastic evolutionary processes.

A novel framework for inferring parameters of transmission from viral sequence data

Transmission between hosts is a critical part of the viral lifecycle. Recent studies of viral transmission have used genome sequence data to evaluate the number of particles transmitted between hosts, and the role of selection as it operates during the transmission process. However, the interpretation of sequence data describing transmission events is a challenging task. We here present a novel and comprehensive framework for using short-read sequence data to understand viral transmission events, designed for influenza virus, but adaptable to other viral species. Our approach solves multiple shortcomings of previous methods for this purpose; for example, we consider transmission as an event involving whole viruses, rather than sets of independent alleles. We demonstrate how selection during transmission and noisy sequence data may each affect naive inferences of the population bottleneck, accounting for these in our framework so as to achieve a correct inference. We identify circumstances in which selection for increased viral transmission may or may not be identified from data. Applying our method to experimental data in which transmission occurs in the presence of strong selection, we show that our framework grants a more quantitative insight into transmission events than previous approaches, inferring the bottleneck in a manner that accounts for selection, both for within-host virulence, and for inherent viral transmissibility. Our work provides new opportunities for studying transmission processes in influenza, and by extension, in other infectious diseases.

Stochastic processes constrain the within and between host evolution of influenza virus

The evolutionary dynamics of influenza virus ultimately derive from processes that take place within and between infected individuals. Here we define influenza virus dynamics in human hosts through sequencing of 249 specimens from 200 individuals collected over 6290 person-seasons of observation. Because these viruses were collected from individuals in a prospective community-based cohort, they are broadly representative of natural infections with seasonal viruses. Consistent with a neutral model of evolution, sequence data from 49 serially sampled individuals illustrated the dynamic turnover of synonymous and nonsynonymous single nucleotide variants and provided little evidence for positive selection of antigenic variants. We also identified 43 genetically-validated transmission pairs in this cohort. Maximum likelihood optimization of multiple transmission models estimated an effective transmission bottleneck of 1-2 genomes. Our data suggest that positive selection is inefficient at the level of the individual host and that stochastic processes dominate the host-level evolution of influenza viruses.

Within-Host Evolution of Human Influenza Virus

The rapid global evolution of influenza virus begins with mutations that arise de novo in individual infections, but little is known about how evolution occurs within hosts. We review recent progress in understanding how and why influenza viruses evolve within human hosts. Advances in deep sequencing make it possible to measure within-host genetic diversity in both acute and chronic influenza infections. Factors like antigenic selection, antiviral treatment, tissue specificity, spatial structure, and multiplicity of infection may affect how influenza viruses evolve within human hosts. Studies of within-host evolution can contribute to our understanding of the evolutionary and epidemiological factors that shape influenza virus's global evolution.

Inferring Fitness Effects from Time-Resolved Sequence Data with a Delay-Deterministic Model

A broad range of approaches have considered the challenge of inferring selection from time-resolved genome sequence data. Models describing deterministic changes in allele or haplotype frequency have been highlighted as providing accurate and computationally...

A common challenge arising from the observation of an evolutionary system over time is to infer the magnitude of selection acting upon a specific genetic variant, or variants, within the population. The inference of selection may be confounded by the effects of genetic drift in a system, leading to the development of inference procedures to account for these effects. However, recent work has suggested that deterministic models of evolution may be effective in capturing the effects of selection even under complex models of demography, suggesting the more general application of deterministic approaches to inference. Responding to this literature, we here note a case in which a deterministic model of evolution may give highly misleading inferences, resulting from the nondeterministic properties of mutation in a finite population. We propose an alternative approach that acts to correct for this error, and which we denote the delay-deterministic model. Applying our model to a simple evolutionary system, we demonstrate its performance in quantifying the extent of selection acting within that system. We further consider the application of our model to sequence data from an evolutionary experiment. We outline scenarios in which our model may produce improved results for the inference of selection, noting that such situations can be easily identified via the use of a regular deterministic model.

Predictive Modeling of Influenza Shows the Promise of Applied Evolutionary Biology

Seasonal influenza is controlled through vaccination campaigns. Evolution of influenza virus antigens means that vaccines must be updated to match novel strains, and vaccine effectiveness depends on the ability of scientists to predict nearly a year in advance which influenza variants will dominate in upcoming seasons. In this review, we highlight a promising new surveillance tool: predictive models. Based on data-sharing and close collaboration between the World Health Organization and academic scientists, these models use surveillance data to make quantitative predictions regarding influenza evolution. Predictive models demonstrate the potential of applied evolutionary biology to improve public health and disease control. We review the state of influenza predictive modeling and discuss next steps and recommendations to ensure that these models deliver upon their considerable biomedical promise.

On the effective depth of viral sequence data

Genome sequence data are of great value in describing evolutionary processes in viral populations. However, in such studies, the extent to which data accurately describes the viral population is a matter of importance. Multiple factors may influence the accuracy of a dataset, including the quantity and nature of the sample collected, and the subsequent steps in viral processing. To investigate this phenomenon, we sequenced replica datasets spanning a range of viruses, and in which the point at which samples were split was different in each case, from a dataset in which independent samples were collected from a single patient to another in which all processing steps up to sequencing were applied to a single sample before splitting the sample and sequencing each replicate. We conclude that neither a high read depth nor a high template number in a sample guarantee the precision of a dataset. Measures of consistency calculated from within a single biological sample may also be insufficient; distortion of the composition of a population by the experimental procedure or genuine within-host diversity between samples may each affect the results. Where it is possible, data from replicate samples should be collected to validate the consistency of short-read sequence data.

The effective rate of influenza reassortment is limited during human infection

We characterise the evolutionary dynamics of influenza infection described by viral sequence data collected from two challenge studies conducted in human hosts. Viral sequence data were collected at regular intervals from infected hosts. Changes in the sequence data observed across time show that the within-host evolution of the virus was driven by the reversion of variants acquired during previous passaging of the virus. Treatment of some patients with oseltamivir on the first day of infection did not lead to the emergence of drug resistance variants in patients. Using an evolutionary model, we inferred the effective rate of reassortment between viral segments, measuring the extent to which randomly chosen viruses within the host exchange genetic material. We find strong evidence that the rate of effective reassortment is low, such that genetic associations between polymorphic loci in different segments are preserved during the course of an infection in a manner not compatible with epistasis. Combining our evidence with that of previous studies we suggest that spatial heterogeneity in the viral population may reduce the extent to which reassortment is observed. Our results do not contradict previous findings of high rates of viral reassortment in vitro and in small animal studies, but indicate that in human hosts the effective rate of reassortment may be substantially more limited.

The influenza virus is an important cause of disease in the human population. During the course of an infection the virus can evolve rapidly. An important mechanism of viral evolution is reassortment, whereby different segments of the influenza genome are shuffled with other segments, producing new viral combinations. Here we study natural selection and reassortment during the course of infections occurring in human hosts. Examining viral genome sequence data from these infections, we note that genetic variants that were acquired during the growth of viruses in culture are selected against in the human host. In addition, we find evidence that the effective rate of reassortment is low. We suggest that the spatial separation between viruses in different parts of the host airway may limit the extent to which genetically distinct segments reassort with one another. Within the global population of influenza viruses, reassortment remains an important factor. However, reassortment is not so rapid as to exclude the possibility of interactions between genome segments affecting the course of influenza evolution during a single infection.

Genetic and antigenic characterization of hemagglutinin of influenza A/H3N2 virus from the 2015 season in Thailand

Antigenic changes in the HA1 domain of the influenza A/H3N2 hemagglutinin (HA) present a challenge in the design of the annual influenza vaccine. We examined the genetic variability in the nucleotide and amino acid of encoding HA1 sequences of the influenza A/H3N2 virus during the 2015 influenza season in Thailand. Toward this, the HA genes of 45 influenza A/H3N2 strains were amplified and sequenced. Although a clade 3C.3a strain (A/Switzerland/9715293/2013) was chosen for the 2015 vaccine, phylogenetic analysis demonstrated that strains belonging to clade 3C.2a (96 %) instead of clade 3C.3a (4 %) were circulating that year. Sequence analysis showed that seven codons were under positive selection, five of which were located inside the antigenic epitopes. The percentages of the perfect match vaccine efficacy (VE) estimated by the P epitope model against circulating strains suggested antigenic drift of the dominant epitopes A and B, which contributed to reduced VE of the 2015 vaccine. However, the 2016 vaccine strain (A/Hong Kong/4801/2014) was closely related and well matched against the circulating strain (mean of VE = 79.3 %). These findings provide data on the antigenic drift of the influenza A/H3N2 virus circulating in Thailand and further support continual monitoring and surveillance of the antigenic changes on HA1.

A conditional likelihood is required to estimate the selection coefficient in ancient DNA

Time-series of allele frequencies are a useful and unique set of data to determine the strength of natural selection on the background of genetic drift. Technically, the selection coefficient is estimated by means of a likelihood function built under the hypothesis that the available trajectory spans a sufficiently large portion of the fitness landscape. Especially for ancient DNA, however, often only one single such trajectories is available and the coverage of the fitness landscape is very limited. In fact, one single trajectory is more representative of a process conditioned both in the initial and in the final condition than of a process free to visit the available fitness landscape. Based on two models of population genetics, here we show how to build a likelihood function for the selection coefficient that takes the statistical peculiarity of single trajectories into account. We show that this conditional likelihood delivers a precise estimate of the selection coefficient also when allele frequencies are close to fixation whereas the unconditioned likelihood fails. Finally, we discuss the fact that the traditional, unconditioned likelihood always delivers an answer, which is often unfalsifiable and appears reasonable also when it is not correct.

Influenza evolution and H3N2 vaccine effectiveness, with application to the 2014/2015 season

Influenza A is a serious disease that causes significant morbidity and mortality, and vaccines against the seasonal influenza disease are of variable effectiveness. In this article, we discuss the use of the pepitope method to predict the dominant influenza strain and the expected vaccine effectiveness in the coming flu season. We illustrate how the effectiveness of the 2014/2015 A/Texas/50/2012 [clade 3C.1] vaccine against the A/California/02/2014 [clade 3C.3a] strain that emerged in the population can be estimated via pepitope In addition, we show by a multidimensional scaling analysis of data collected through 2014, the emergence of a new A/New Mexico/11/2014-like cluster [clade 3C.2a] that is immunologically distinct from the A/California/02/2014-like strains.

SAMFIRE: multi-locus variant calling for time-resolved sequence data

An increasingly common method for studying evolution is the collection of time-resolved short-read sequence data. Such datasets allow for the direct observation of rapid evolutionary processes, as might occur in natural microbial populations and in evolutionary experiments. In many circumstances, evolutionary pressure acting upon single variants can cause genomic changes at multiple nearby loci. SAMFIRE is an open-access software package for processing and analyzing sequence reads from time-resolved data, calling important single- and multi-locus variants over time, identifying alleles potentially affected by selection, calculating linkage disequilibrium statistics, performing haplotype reconstruction and exploiting time-resolved information to estimate the extent of uncertainty in reported genomic data.

AVAILABILITY AND IMPLEMENTATION

C ++ code may be found at https://github.com/cjri/samfire/

CONTACT

chris.illingworth@gen.cam.ac.uk

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Deep Sequencing Reveals Potential Antigenic Variants at Low Frequencies in Influenza A Virus-Infected Humans

Influenza vaccines must be frequently reformulated to account for antigenic changes in the viral envelope protein, hemagglutinin (HA). The rapid evolution of influenza virus under immune pressure is likely enhanced by the virus's genetic diversity within a host, although antigenic change has rarely been investigated on the level of individual infected humans. We used deep sequencing to characterize the between- and within-host genetic diversity of influenza viruses in a cohort of patients that included individuals who were vaccinated and then infected in the same season. We characterized influenza HA segments from the predominant circulating influenza A subtypes during the 2012-2013 (H3N2) and 2013-2014 (pandemic H1N1; H1N1pdm) flu seasons. We found that HA consensus sequences were similar in nonvaccinated and vaccinated subjects. In both groups, purifying selection was the dominant force shaping HA genetic diversity. Interestingly, viruses from multiple individuals harbored low-frequency mutations encoding amino acid substitutions in HA antigenic sites at or near the receptor-binding domain. These mutations included two substitutions in H1N1pdm viruses, G158K and N159K, which were recently found to confer escape from virus-specific antibodies. These findings raise the possibility that influenza antigenic diversity can be generated within individual human hosts but may not become fixed in the viral population even when they would be expected to have a strong fitness advantage. Understanding constraints on influenza antigenic evolution within individual hosts may elucidate potential future pathways of antigenic evolution at the population level.

IMPORTANCE

Influenza vaccines must be frequently reformulated due to the virus's rapid evolution rate. We know that influenza viruses exist within each infected host as a "swarm" of genetically distinct viruses, but the role of this within-host diversity in the antigenic evolution of influenza has been unclear. We characterized here the genetic and potential antigenic diversity of influenza viruses infecting humans, some of whom became infected despite recent vaccination. Influenza virus between- and within-host genetic diversity was not significantly different in nonvaccinated and vaccinated humans, suggesting that vaccine-induced immunity does not exert strong selective pressure on viruses replicating in individual people. We found low-frequency mutations, below the detection threshold of traditional surveillance methods, in nonvaccinated and vaccinated humans that were recently associated with antibody escape. Interestingly, these potential antigenic variants did not reach fixation in infected people, suggesting that other evolutionary factors may be hindering their emergence in individual humans.

Fitness Inference from Short-Read Data: Within-Host Evolution of a Reassortant H5N1 Influenza Virus

We present a method to infer the role of selection acting during the within-host evolution of the influenza virus from short-read genome sequence data. Linkage disequilibrium between loci is accounted for by treating short-read sequences as noisy multilocus emissions from an underlying model of haplotype evolution. A hierarchical model-selection procedure is used to infer the underlying fitness landscape of the virus insofar as that landscape is explored by the viral population. In a first application of our method, we analyze data from an evolutionary experiment describing the growth of a reassortant H5N1 virus in ferrets. Across two sets of replica experiments we infer multiple alleles to be under selection, including variants associated with receptor binding specificity, glycosylation, and with the increased transmissibility of the virus. We identify epistasis as an important component of the within-host fitness landscape, and show that adaptation can proceed through multiple genetic pathways.

Improving pandemic influenza risk assessment

Assessing the pandemic risk posed by specific non-human influenza A viruses is an important goal in public health research. As influenza virus genome sequencing becomes cheaper, faster, and more readily available, the ability to predict pandemic potential from sequence data could transform pandemic influenza risk assessment capabilities. However, the complexities of the relationships between virus genotype and phenotype make such predictions extremely difficult. The integration of experimental work, computational tool development, and analysis of evolutionary pathways, together with refinements to influenza surveillance, has the potential to transform our ability to assess the risks posed to humans by non-human influenza viruses and lead to improved pandemic preparedness and response.

Population genetics inference for longitudinally-sampled mutants under strong selection

Longitudinal allele frequency data are becoming increasingly prevalent. Such samples permit statistical inference of the population genetics parameters that influence the fate of mutant variants. To infer these parameters by maximum likelihood, the mutant frequency is often assumed to evolve according to the Wright–Fisher model. For computational reasons, this discrete model is commonly approximated by a diffusion process that requires the assumption that the forces of natural selection and mutation are weak. This assumption is not always appropriate. For example, mutations that impart drug resistance in pathogens may evolve under strong selective pressure. Here, we present an alternative approximation to the mutant-frequency distribution that does not make any assumptions about the magnitude of selection or mutation and is much more computationally efficient than the standard diffusion approximation. Simulation studies are used to compare the performance of our method to that of the Wright–Fisher and Gaussian diffusion approximations. For large populations, our method is found to provide a much better approximation to the mutant-frequency distribution when selection is strong, while all three methods perform comparably when selection is weak. Importantly, maximum-likelihood estimates of the selection coefficient are severely attenuated when selection is strong under the two diffusion models, but not when our method is used. This is further demonstrated with an application to mutant-frequency data from an experimental study of bacteriophage evolution. We therefore recommend our method for estimating the selection coefficient when the effective population size is too large to utilize the discrete Wright–Fisher model.