Effects of passage history and sampling bias on phylogenetic reconstruction of human influenza A evolution

Investigation of cross-regional spread and evolution of equine influenza H3N8 at US and global scales using Bayesian phylogeography based on balanced subsampling

Equine influenza virus (EIV) is a highly contagious pathogen of equids, and a well-known burden in global equine health. EIV H3N8 variants seasonally emerged and resulted in EIV outbreaks in the United States (US) and worldwide. The present study evaluated the pattern of cross-regional EIV H3N8 spread and evolutionary characteristics at US and global scales using Bayesian phylogeography with balanced subsampling based on regional horse population size. A total of 297 Haemagglutinin (HA) sequences of global EIV H3N8 were collected from 1963 to 2019 and subsampled to global subset (n = 67), raw US sequences (n = 100) and US subset (n = 44) datasets. Discrete trait phylogeography analysis was used to estimate the transmission history of EIV using four global and US genome datasets. The North American lineage was the major source of globally dominant EIV variants and spread to other global regions. The US EIV strains generally spread from the southern and midwestern regions to other regions. The EIV H3N8 accumulated approximately three nucleotide substitutions per year in the HA gene under heterogenous local positive selection. Our findings will guide better decision making of target intervention strategies of EIV H3N8 infection and provide the better scheme of genomic surveillance in the US and global equine health. This article is protected by copyright. All rights reserved.

Passage Adaptation Correlates With the Reduced Efficacy of the Influenza Vaccine

BACKGROUND

As a dominant seasonal influenza virus, H3N2 virus rapidly evolves in humans and is a constant threat to public health. Despite sustained research efforts, the efficacy of H3N2 vaccine has decreased rapidly. Even though antigenic drift and passage adaptation (substitutions accumulated during vaccine production in embryonated eggs) have been implicated in reduced vaccine efficacy (VE), their respective contributions to the phenomenon remain controversial.

METHODS

We utilized mutational mapping, a powerful probabilistic method for studying sequence evolution, to analyze patterns of substitutions in different passage conditions for an unprecedented amount of H3N2 hemagglutinin sequences (n = 32 278).

RESULTS

We found that passage adaptation in embryonated eggs is driven by repeated convergent evolution over 12 codons. Based on substitution patterns at these sites, we developed a metric, adaptive distance (AD), to quantify the strength of passage adaptation and subsequently identified a strong negative correlation between AD and VE.

CONCLUSIONS

The high correlation between AD and VE implies that passage adaptation in embryonated eggs may be a strong contributor to the recent reduction in H3N2 VE. We developed a computational package called MADE (Measuring Adaptive Distance and vaccine Efficacy based on allelic barcodes) to measure the strength of passage adaptation and predict the efficacy of a candidate vaccine strain. Our findings shed light on strategies for reducing Darwinian evolution within the passaging medium in order to potentially restore an effective vaccine program in the future.

Population genomic transformations induced by isolation of wild bird avian influenza viruses (Orthomyxoviridae) in embryonated chicken eggs

Isolation and cultivation of wild-type viruses in model organism cells or tissues is standard practice in virology. Oftentimes, the virus host species is distantly related to the species from which the culture system was developed. Thus, virus culture in these tissues and cells basically constitutes a host jump, which can lead to genomic changes through genetic drift and/or adaptation to the culture system. We directly sequenced 70 avian influenza virus (Orthomyxoviridae) genomes from oropharyngeal/cloacal swabs collected from wild bird species and paired virus isolates propagated from the same samples following isolation in specific-pathogen-free embryonated chicken eggs. The data were analyzed using population genetic approaches including evaluation of single nucleotide polymorphism (SNP) frequencies and divergence with pooled-sequencing analyses, consensus sequence placement in neighbor-joining trees, and haplotype reconstruction and networks. We found that propagation of virus in eggs leads to skewed SNP mutation spectra with some SNPs going to fixation. Both synonymous and nonsynonmous SNP frequencies shifted. We found multiple consensus sequences that differed between the swabs and the isolates, with some sequences from the same sample falling into divergent genetic clusters. Twenty of 23 coinfections detected had different dominant subtypes following virus isolation, thus sequences from both the swab and isolate were needed to obtain full subtype data. Haplotype networks revealed haplotype frequency shifts and the appearance or loss of low-frequency haplotypes following isolation. The results from this study revealed that isolation of wild bird avian influenza viruses in chicken eggs leads to skewed populations that are different than the input populations. Consensus sequence changes from virus isolation can lead to flawed phylogenetic inferences, and subtype detection is biased. These results suggest that for genomic studies of wild bird influenza viruses the biological field should move away from chicken egg isolation towards directly sequencing the virus from host samples.

Allele-specific nonstationarity in evolution of influenza A virus surface proteins

Influenza A virus (IAV) is a major public health problem and a pandemic threat. Its evolution is largely driven by diversifying positive selection so that relative fitness of different amino acid variants changes with time due to changes in herd immunity or genomic context, and novel amino acid variants attain fitness advantage. Here, we hypothesize that diversifying selection also has another manifestation: the fitness associated with a particular amino acid variant should decline with time since its origin, as the herd immunity adapts to it. By tracing the evolution of antigenic sites at IAV surface proteins, we show that an amino acid variant becomes progressively more likely to become replaced by another variant with time since its origin-a phenomenon we call "senescence." Senescence is particularly pronounced at experimentally validated antigenic sites, implying that it is largely driven by host immunity. By contrast, at internal sites, existing variants become more favorable with time, probably due to arising contingent mutations at other epistatically interacting sites. Our findings reveal a previously undescribed facet of adaptive evolution and suggest approaches for prediction of evolutionary dynamics of pathogens.

Influenza A Hemagglutinin Passage Bias Sites and Host Specificity Mutations

Animal studies aimed at understanding influenza virus mutations that change host specificity to adapt to replication in mammalian hosts are necessarily limited in sample numbers due to high cost and safety requirements. As a safe, higher-throughput alternative, we explore the possibility of using readily available passage bias data obtained mostly from seasonal H1 and H3 influenza strains that were differentially grown in mammalian (MDCK) and avian cells (eggs). Using a statistical approach over 80,000 influenza hemagglutinin sequences with passage information, we found that passage bias sites are most commonly found in three regions: (i) the globular head domain around the receptor binding site, (ii) the region that undergoes pH-dependent structural changes and (iii) the unstructured N-terminal region harbouring the signal peptide. Passage bias sites were consistent among different passage cell types as well as between influenza A subtypes. We also find epistatic interactions of site pairs supporting the notion of host-specific dependency of mutations on virus genomic background. The sites identified from our large-scale sequence analysis substantially overlap with known host adaptation sites in the WHO H5N1 genetic changes inventory suggesting information from passage bias can provide candidate sites for host specificity changes to aid in risk assessment for emerging strains.

Deep sequencing of 2009 influenza A/H1N1 virus isolated from volunteer human challenge study participants and natural infections

Nasal wash samples from 15 human volunteers challenged with GMP manufactured influenza A/California/04/2009(H1N1) and from 5 naturally infected influenza patients of the 2009 pandemic were deep sequenced using viral targeted hybridization enrichment. Ten single nucleotide polymorphism (SNP) positions were found in the challenge virus. Some of the nonsynonymous changes in the inoculant virus were maintained in some challenge participants, but not in others, indicating that virus is evolving away from the Vero cell adapted inoculant, for example SNPs in the neuraminidase. Many SNP sites in challenge patients and naturally infected patients were found, many not identified previously. The SNPs identified, and phylogenetic analyses, showed that intrahost evolution of the virus are different in challenge participants and naturally infected patients. This study, using hybridization enrichment without PCR, provided an accurate and unbiased assessment of differential intrahost viral evolution from a uniform influenza inoculant in humans and comparison to naturally infected patients.

In silico re-assessment of a diagnostic RT-qPCR assay for universal detection of Influenza A viruses

The ongoing evolution of microbial pathogens represents a significant issue in diagnostic PCR/qPCR. Many assays are burdened with false negativity due to mispriming and/or probe-binding failures. Therefore, PCR/qPCR assays used in the laboratory should be periodically re-assessed in silico on public sequences to evaluate the ability to detect actually circulating strains and to infer potentially escaping variants. In the work presented we re-assessed a RT-qPCR assay for the universal detection of influenza A (IA) viruses currently recommended by the European Union Reference Laboratory for Avian Influenza. To this end, the primers and probe sequences were challenged against more than 99,000 M-segment sequences in five data pools. To streamline this process, we developed a simple algorithm called the SequenceTracer designed for alignment stratification, compression, and personal sequence subset selection and also demonstrated its utility. The re-assessment confirmed the high inclusivity of the assay for the detection of avian, swine and human pandemic H1N1 IA viruses. On the other hand, the analysis identified human H3N2 strains with a critical probe-interfering mutation circulating since 2010, albeit with a significantly fluctuating proportion. Minor variations located in the forward and reverse primers identified in the avian and swine data were also considered.

Influenza passaging annotations: what they tell us and why we should listen

Influenza databases now contain over 100,000 worldwide sequence records for strains influenza A(H3N2) and A(H1N1). Although these data facilitate global research efforts and vaccine development practices, they also represent a stumbling block for researchers because of their confusing and heterogeneous annotation. Unclear passaging annotations are particularly concerning given the recent work highlighting the presence and risk of false adaptation signals introduced by cell passaging of viral isolates. With this in mind, we aim to provide a concise outline of why viruses are passaged, a clear overview of passaging annotation nomenclature currently in use, and suggestions for a standardized nomenclature going forward. Our hope is that this summary will empower researchers and clinicians alike to more easily understand a virus sample's passage history when analyzing influenza sequences.

Adaptive Evolution of Human-Isolated H5Nx Avian Influenza A Viruses

Avian influenza A viruses (AIVs) H5N1, first identified in 1996, are highly pathogenic in domestic poultry and continue to occasionally infect humans. In this study, we sought to identify genetic changes that occurred during their multiple invasions to humans. We evaluated all available H5Nx AIV genomes. Significant signals of positive selection were detected in 29 host-shift branches. 126 parallel evolution sites were detected on these branches, including 17 well-known sites (such as T271A, A274T, T339M, Q591K, E627K, and D701N in PB2; A134V, D154N, S223N, and R497K in HA) that play roles in allowing AIVs to cross species barriers. Our study suggests that during human infections, H5Nx viruses have experienced adaptive evolution (positive selection and convergent evolution) that allowed them to adapt to their new host environments. Analyses of adaptive evolution should be useful in identifying candidate sites that play roles in human infections, which can be tested by functional experiments.

The evolutionary dynamics of H1N1/pdm2009 in India

After the 2009 pandemic H1N1 outbreak, this new reassortant virus replaced the previous seasonal H1N1 and established itself in human populations. In 2014-2015, the H1N1/pdm2009 pandemic strain caused the worst influenza outbreak in India in recent years. It is still unclear whether this outbreak was due to a genetic change in the H1N1/pdm2009 virus. Here, we collected all available data from India, and used data from other South and Southeast Asian countries (Bangladesh, Myanmar, Nepal, Singapore, and Vietnam) that are geographically nearby India for comparison, to examine the evolutionary dynamics of this influenza virus. We detected signals of population expansion in both HA and NA in the 2014/2015 season in India, which coincided with the outbreak in India of 2015. Signals of positive selection in HA in the 2012/2013 and 2013/2014 seasons, and in NA in the 2012/2013 and 2014/2015 seasons were also found. These selective patterns were different from those seen in other South and Southeast Asian countries. However, none of the eight positively selected sites are endemic to India. The rapid circulation of H1N1/pdm2009 between these South and Southeast Asian countries may explain this result. Further studies are still needed to identify the genetic basis of the great 2015 outbreak in India.

Viral factors in influenza pandemic risk assessment

The threat of an influenza A virus pandemic stems from continual virus spillovers from reservoir species, a tiny fraction of which spark sustained transmission in humans. To date, no pandemic emergence of a new influenza strain has been preceded by detection of a closely related precursor in an animal or human. Nonetheless, influenza surveillance efforts are expanding, prompting a need for tools to assess the pandemic risk posed by a detected virus. The goal would be to use genetic sequence and/or biological assays of viral traits to identify those non-human influenza viruses with the greatest risk of evolving into pandemic threats, and/or to understand drivers of such evolution, to prioritize pandemic prevention or response measures. We describe such efforts, identify progress and ongoing challenges, and discuss three specific traits of influenza viruses (hemagglutinin receptor binding specificity, hemagglutinin pH of activation, and polymerase complex efficiency) that contribute to pandemic risk.

Dynamic Convergent Evolution Drives the Passage Adaptation across 48 Years' History of H3N2 Influenza Evolution

Influenza viruses are often propagated in a diverse set of culturing media and additional substitutions known as passage adaptation can cause extra evolution in the target strain, leading to ineffective vaccines. Using 25,482 H3N2 HA1 sequences curated from Global Initiative on Sharing All Influenza Data and National Center for Biotechnology Information databases, we found that passage adaptation is a very dynamic process that changes over time and evolves in a seesaw like pattern. After crossing the species boundary from bird to human in 1968, the influenza H3N2 virus evolves to be better adapted to the human environment and passaging them in embryonated eggs (i.e., an avian environment) leads to increasingly stronger positive selection. On the contrary, passage adaptation to the mammalian cell lines changes from positive selection to negative selection. Using two statistical tests, we identified 19 codon positions around the receptor binding domain strongly contributing to passage adaptation in the embryonated egg. These sites show strong convergent evolution and overlap extensively with positively selected sites identified in humans, suggesting that passage adaptation can confound many of the earlier studies on influenza evolution. Interestingly, passage adaptation in recent years seems to target a few codon positions in antigenic surface epitopes, which makes it difficult to produce antigenically unaltered vaccines using embryonic eggs. Our study outlines another interesting scenario whereby both convergent and adaptive evolution are working in synchrony driving viral adaptation. Future studies from sequence analysis to vaccine production need to take careful consideration of passage adaptation.

Sequence amplification via cell passaging creates spurious signals of positive adaptation in influenza virus H3N2 hemagglutinin

Clinical influenza A virus isolates are frequently not sequenced directly. Instead, a majority of these isolates (~70% in 2015) are first subjected to passaging for amplification, most commonly in non-human cell culture. Here, we find that this passaging leaves distinct signals of adaptation, which can confound evolutionary analyses of the viral sequences. We find distinct patterns of adaptation to Madin-Darby (MDCK) and monkey cell culture absent from unpassaged hemagglutinin sequences. These patterns also dominate pooled datasets not separated by passaging type, and they increase in proportion to the number of passages performed. By contrast, MDCK-SIAT1 passaged sequences seem mostly (but not entirely) free of passaging adaptations. Contrary to previous studies, we find that using only internal branches of influenza virus phylogenetic trees is insufficient to correct for passaging artifacts. These artifacts can only be safely avoided by excluding passaged sequences entirely from subsequent analysis. We conclude that future influenza virus evolutionary analyses should appropriately control for potentially confounding effects of passaging adaptations.

H3N2 Mismatch of 2014-15 Northern Hemisphere Influenza Vaccines and Head-to-head Comparison between Human and Ferret Antisera derived Antigenic Maps

The poor performance of 2014–15 Northern Hemisphere (NH) influenza vaccines was attributed to mismatched H3N2 component with circulating epidemic strains. Using human serum samples collected from 2009–10, 2010–11 and 2014–15 NH influenza vaccine trials, we assessed their cross-reactive hemagglutination inhibition (HAI) antibody responses against recent H3 epidemic isolates. All three populations (children, adults, and older adults) vaccinated with the 2014–15 NH egg- or cell-based vaccine, showed >50% reduction in HAI post-vaccination geometric mean titers against epidemic H3 isolates from those against egg-grown H3 vaccine strain A/Texas/50/2012 (TX/12e). The 2014–15 NH vaccines, regardless of production type, failed to further extend HAI cross-reactivity against H3 epidemic strains from previous seasonal vaccines. Head-to-head comparison between ferret and human antisera derived antigenic maps revealed different antigenic patterns among representative egg- and cell-grown H3 viruses characterized. Molecular modeling indicated that the mutations of epidemic H3 strains were mainly located in antibody-binding sites A and B as compared with TX/12e. To improve vaccine strain selection, human serologic testing on vaccination-induced cross-reactivity need be emphasized along with virus antigenic characterization by ferret model.

Dynamically correlated mutations drive human Influenza A evolution

Human Influenza A virus undergoes recurrent changes in the hemagglutinin (HA) surface protein, primarily involved in the human antibody recognition. Relevant antigenic changes, enabling the virus to evade host immune response, have been recognized to occur in parallel to multiple mutations at antigenic sites in HA. Yet, the role of correlated mutations (epistasis) in driving the molecular evolution of the virus still represents a challenging puzzle. Further, though circulation at a global geographic level is key for the survival of Influenza A, its role in shaping the viral phylodynamics remains largely unexplored. Here we show, through a sequence based epidemiological model, that epistatic effects between amino acids substitutions, coupled with a reservoir that mimics worldwide circulating viruses, are key determinants that drive human Influenza A evolution. Our approach explains all the up-to-date observations characterizing the evolution of H3N2 subtype, including phylogenetic properties, nucleotide fixation patterns, and composition of antigenic clusters.

A predictive fitness model for influenza

The seasonal human influenza A/H3N2 virus undergoes rapid evolution, which produces significant year-to-year sequence turnover in the population of circulating strains. Adaptive mutations respond to human immune challenge and occur primarily in antigenic epitopes, the antibody-binding domains of the viral surface protein haemagglutinin. Here we develop a fitness model for haemagglutinin that predicts the evolution of the viral population from one year to the next. Two factors are shown to determine the fitness of a strain: adaptive epitope changes and deleterious mutations outside the epitopes. We infer both fitness components for the strains circulating in a given year, using population-genetic data of all previous strains. From fitness and frequency of each strain, we predict the frequency of its descendent strains in the following year. This fitness model maps the adaptive history of influenza A and suggests a principled method for vaccine selection. Our results call for a more comprehensive epidemiology of influenza and other fast-evolving pathogens that integrates antigenic phenotypes with other viral functions coupled by genetic linkage.

Comparison of mutation patterns in full-genome A/H3N2 influenza sequences obtained directly from clinical samples and the same samples after a single MDCK passage

Human influenza viruses can be isolated efficiently from clinical samples using Madin-Darby canine kidney (MDCK) cells. However, this process is known to induce mutations in the virus as it adapts to this non-human cell-line. We performed a systematic study to record the pattern of MDCK-induced mutations observed across the whole influenza A/H3N2 genome. Seventy-seven clinical samples collected from 2009-2011 were included in the study. Two full influenza genomes were obtained for each sample: one from virus obtained directly from the clinical sample and one from the matching isolate cultured in MDCK cells. Comparison of the full-genome sequences obtained from each of these sources showed that 42% of the 77 isolates had acquired at least one MDCK-induced mutation. The presence or absence of these mutations was independent of viral load or sample origin (in-patients versus out-patients). Notably, all the five hemagglutinin missense mutations were observed at the hemaggutinin 1 domain only, particularly within or proximal to the receptor binding sites and antigenic site of the virus. Furthermore, 23% of the 77 isolates had undergone a MDCK-induced missense mutation, D151G/N, in the neuraminidase segment. This mutation has been found to be associated with reduced drug sensitivity towards the neuraminidase inhibitors and increased viral receptor binding efficiency to host cells. In contrast, none of the neuraminidase sequences obtained directly from the clinical samples contained the D151G/N mutation, suggesting that this mutation may be an indicator of MDCK culture-induced changes. These D151 mutations can confound the interpretation of the hemagglutination inhibition assay and neuraminidase inhibitor resistance results when these are based on MDCK isolates. Such isolates are currently in routine use in the WHO influenza vaccine and drug-resistance surveillance programs. Potential data interpretation miscalls can therefore be avoided by careful exclusion of such D151 mutants after further sequence analysis.

Simplified large-scale Sanger genome sequencing for influenza A/H3N2 virus

Background

The advent of next-generation sequencing technologies and the resultant lower costs of sequencing have enabled production of massive amounts of data, including the generation of full genome sequences of pathogens. However, the small genome size of the influenza virus arguably justifies the use of the more conventional Sanger sequencing technology which is still currently more readily available in most diagnostic laboratories.

Results

We present a simplified Sanger-based genome sequencing method for sequencing the influenza A/H3N2 virus in a large-scale format. The entire genome sequencing was completed with 19 reverse transcription-polymerase chain reactions (RT-PCRs) and 39 sequencing reactions. This method was tested on 15 native clinical samples and 15 culture isolates, respectively, collected between 2009 and 2011. The 15 native clinical samples registered quantification cycle values ranging from 21.0 to 30.56, which were equivalent to 2.4×10³–1.4×10⁶ viral copies/µL of RNA extract. All the PCR-amplified products were sequenced directly without PCR product purification. Notably, high quality sequencing data up to 700 bp were generated for all the samples tested. The completed sequence covered 408,810 nucleotides in total, with 13,627 nucleotides per genome, attaining 100% coding completeness. Of all the bases produced, an average of 89.49% were Phred quality value 40 (QV40) bases (representing an accuracy of circa one miscall for every 10,000 bases) or higher, and an average of 93.46% were QV30 bases (one miscall every 1000 bases) or higher.

Conclusions

This sequencing protocol has been shown to be cost-effective and less labor-intensive in obtaining full influenza genomes. The constant high quality of sequences generated imparts confidence in extending the application of this non-purified amplicon sequencing approach to other gene sequencing assays, with appropriate use of suitably designed primers.

Effect of long-term laboratory propagation on Chlamydia trachomatis genome dynamics

It is assumed that bacterial strains maintained in the laboratory for long time shape their genome in a different fashion from the nature-circulating strains. Here, we analyzed the impact of long-term in vitro propagation on the genome of the obligate intracellular pathogen Chlamydia trachomatis. We fully-sequenced the genome of a historical prototype strain (L2/434/Bu) and a clinical isolate (E/CS88), before and after one-year of serial in vitro passaging (up to 3500 bacterial generations). We observed a slow adaptation of C. trachomatis to the in vitro environment, which was essentially governed by four mutations for L2/434/Bu and solely one mutation for E/CS88, corresponding to estimated mutation rates from 3.84 × 10(-10) to 1.10 × 10(-9) mutations per base pair per generation. In a speculative basis, the mutations likely conferred selective advantage as: (i) mathematical modeling showed that selective advantage is mandatory for frequency increase of a mutated clone; (ii) transversions and non-synonymous mutations were overrepresented; (iii) two non-synonymous mutations affected the genes CTL0084 and CTL0610, encoding a putative transferase and a protein likely implicated in transcription regulation respectively, which are families known to be highly prone to undergone laboratory-derived advantageous mutations in other bacteria; and (iv) the mutation for E/CS88 is located likely in the regulatory region of a virulence gene (CT115/incD) believed to play a role in subverting the host cell machinery. Nevertheless, we found no significant differences in the growth rate, plasmid load, and attachment/entry rate, between strains before and after their long-term laboratory propagation. Of note, from the mixture of clones in E/CS88 initial population, an inactivating mutation in the virulence gene CT135 evolved to 100% prevalence, unequivocally indicating that this gene is superfluous for C. trachomatis survival in vitro. Globally, C. trachomatis revealed a slow in vitro adaptation that only modestly modifies the in vivo-derived genomic evolutionary landscape.

Irreconcilable differences: divorcing geographic mutation and recombination rates within a global MRSA clone

A growing resource of methicillin-resistant Staphylococcus aureus (MRSA) genomes uncovers intriguing phylogeographic and recombination patterns and highlights challenges in identifying the source of these phenomena.

Phylogenetic properties of RNA viruses

A new word, phylodynamics, was coined to emphasize the interconnection between phylogenetic properties, as observed for instance in a phylogenetic tree, and the epidemic dynamics of viruses, where selection, mediated by the host immune response, and transmission play a crucial role. The challenges faced when investigating the evolution of RNA viruses call for a virtuous loop of data collection, data analysis and modeling. This already resulted both in the collection of massive sequences databases and in the formulation of hypotheses on the main mechanisms driving qualitative differences observed in the (reconstructed) evolutionary patterns of different RNA viruses. Qualitatively, it has been observed that selection driven by the host immune response induces an uneven survival ability among co-existing strains. As a consequence, the imbalance level of the phylogenetic tree is manifestly more pronounced if compared to the case when the interaction with the host immune system does not play a central role in the evolutive dynamics. While many imbalance metrics have been introduced, reliable methods to discriminate in a quantitative way different level of imbalance are still lacking. In our work, we reconstruct and analyze the phylogenetic trees of six RNA viruses, with a special emphasis on the human Influenza A virus, due to its relevance for vaccine preparation as well as for the theoretical challenges it poses due to its peculiar evolutionary dynamics. We focus in particular on topological properties. We point out the limitation featured by standard imbalance metrics, and we introduce a new methodology with which we assign the correct imbalance level of the phylogenetic trees, in agreement with the phylodynamics of the viruses. Our thorough quantitative analysis allows for a deeper understanding of the evolutionary dynamics of the considered RNA viruses, which is crucial in order to provide a valuable framework for a quantitative assessment of theoretical predictions.

Fatal cases of influenza A(H3N2) in children: insights from whole genome sequence analysis

During the Northern Hemisphere winter of 2003–2004 the emergence of a novel influenza antigenic variant, A/Fujian/411/2002-like(H3N2), was associated with an unusually high number of fatalities in children. Seventeen fatal cases in the UK were laboratory confirmed for Fujian/411-like viruses. To look for phylogenetic patterns and genetic markers that might be associated with increased virulence, sequencing and phylogenetic analysis of the whole genomes of 63 viruses isolated from fatal cases and non fatal “control” cases was undertaken. The analysis revealed the circulation of two main genetic groups, I and II, both of which contained viruses from fatal cases. No associated amino acid substitutions could be linked with an exclusive or higher occurrence in fatal cases. The Fujian/411-like viruses in genetic groups I and II completely displaced other A(H3N2) viruses, but they disappeared after 2004. This study shows that two A(H3N2) virus genotypes circulated exclusively during the winter of 2003–2004 in the UK and caused an unusually high number of deaths in children. Host factors related to immune state and differences in genetic background between patients may also play important roles in determining the outcome of an influenza infection.

Purifying selection can obscure the ancient age of viral lineages

Statistical methods for molecular dating of viral origins have been used extensively to infer the time of most common recent ancestor for many rapidly evolving pathogens. However, there are a number of cases, in which epidemiological, historical, or genomic evidence suggests much older viral origins than those obtained via molecular dating. We demonstrate how pervasive purifying selection can mask the ancient origins of recently sampled pathogens, in part due to the inability of nucleotide-based substitution models to properly account for complex patterns of spatial and temporal variability in selective pressures. We use codon-based substitution models to infer the length of branches in viral phylogenies; these models produce estimates that are often considerably longer than those obtained with traditional nucleotide-based substitution models. Correcting the apparent underestimation of branch lengths suggests substantially older origins for measles, Ebola, and avian influenza viruses. This work helps to reconcile some of the inconsistencies between molecular dating and other types of evidence concerning the age of viral lineages.

Reconstructing disease outbreaks from genetic data: a graph approach

Epidemiology and public health planning will increasingly rely on the analysis of genetic sequence data. In particular, genetic data coupled with dates and locations of sampled isolates can be used to reconstruct the spatiotemporal dynamics of pathogens during outbreaks. Thus far, phylogenetic methods have been used to tackle this issue. Although these approaches have proved useful for informing on the spread of pathogens, they do not aim at directly reconstructing the underlying transmission tree. Instead, phylogenetic models infer most recent common ancestors between pairs of isolates, which can be inadequate for densely sampled recent outbreaks, where the sample includes ancestral and descendent isolates. In this paper, we introduce a novel method based on a graph approach to reconstruct transmission trees directly from genetic data. Using simulated data, we show that our approach can efficiently reconstruct genealogies of isolates in situations where classical phylogenetic approaches fail to do so. We then illustrate our method by analyzing data from the early stages of the swine-origin A/H1N1 influenza pandemic. Using 433 isolates sequenced at both the hemagglutinin and neuraminidase genes, we reconstruct the likely history of the worldwide spread of this new influenza strain. The presented methodology opens new perspectives for the analysis of genetic data in the context of disease outbreaks.

The re-emergence of H1N1 influenza virus in 1977: a cautionary tale for estimating divergence times using biologically unrealistic sampling dates

In 1977, H1N1 influenza A virus reappeared after a 20-year absence. Genetic analysis indicated that this strain was missing decades of nucleotide sequence evolution, suggesting an accidental release of a frozen laboratory strain into the general population. Recently, this strain and its descendants were included in an analysis attempting to date the origin of pandemic influenza virus without accounting for the missing decades of evolution. Here, we investigated the effect of using viral isolates with biologically unrealistic sampling dates on estimates of divergence dates. Not accounting for missing sequence evolution produced biased results and increased the variance of date estimates of the most recent common ancestor of the re-emergent lineages and across the entire phylogeny. Reanalysis of the H1N1 sequences excluding isolates with unrealistic sampling dates indicates that the 1977 re-emergent lineage was circulating for approximately one year before detection, making it difficult to determine the geographic source of reintroduction. We suggest that a new method is needed to account for viral isolates with unrealistic sampling dates.

Genetic and pathobiologic characterization of pandemic H1N1 2009 influenza viruses from a naturally infected swine herd

Since its initial identification in Mexico and the United States, concerns have been raised that the novel H1N1 influenza virus might cause a pandemic of severity comparable to that of the 1918 pandemic. In late April 2009, viruses phylogenetically related to pandemic H1N1 influenza virus were isolated from an outbreak on a Canadian pig farm. This outbreak also had epidemiological links to a suspected human case. Experimental infections carried out in pigs using one of the swine isolates from this outbreak and the human isolate A/Mexico/InDRE4487/2009 showed differences in virus recovery from the lower respiratory tract. Virus was consistently isolated from the lungs of pigs infected with A/Mexico/InDRE4487/2009, while only one pig infected with A/swine/Alberta/OTH-33-8/2008 yielded live virus from the lung, despite comparable amounts of viral RNA and antigen in both groups of pigs. Clinical disease resembled other influenza virus infections in swine, albeit with somewhat prolonged virus antigen detection and delayed viral-RNA clearance from the lungs. There was also a noteworthy amount of genotypic variability among the viruses isolated from the pigs on the farm. This, along with the somewhat irregular pathobiological characteristics observed in experimentally infected animals, suggests that although the virus may be of swine origin, significant viral evolution may still be ongoing.

From where did the 2009 'swine-origin' influenza A virus (H1N1) emerge?

The swine-origin influenza A (H1N1) virus that appeared in 2009 and was first found in human beings in Mexico, is a reassortant with at least three parents. Six of the genes are closest in sequence to those of H1N2 'triple-reassortant' influenza viruses isolated from pigs in North America around 1999-2000. Its other two genes are from different Eurasian 'avian-like' viruses of pigs; the NA gene is closest to H1N1 viruses isolated in Europe in 1991-1993, and the MP gene is closest to H3N2 viruses isolated in Asia in 1999-2000. The sequences of these genes do not directly reveal the immediate source of the virus as the closest were from isolates collected more than a decade before the human pandemic started. The three parents of the virus may have been assembled in one place by natural means, such as by migrating birds, however the consistent link with pig viruses suggests that human activity was involved. We discuss a published suggestion that unsampled pig herds, the intercontinental live pig trade, together with porous quarantine barriers, generated the reassortant. We contrast that suggestion with the possibility that laboratory errors involving the sharing of virus isolates and cultured cells, or perhaps vaccine production, may have been involved. Gene sequences from isolates that bridge the time and phylogenetic gap between the new virus and its parents will distinguish between these possibilities, and we suggest where they should be sought. It is important that the source of the new virus be found if we wish to avoid future pandemics rather than just trying to minimize the consequences after they have emerged. Influenza virus is a very significant zoonotic pathogen. Public confidence in influenza research, and the agribusinesses that are based on influenza's many hosts, has been eroded by several recent events involving the virus. Measures that might restore confidence include establishing a unified international administrative framework coordinating surveillance, research and commercial work with this virus, and maintaining a registry of all influenza isolates.

Using non-homogeneous models of nucleotide substitution to identify host shift events: application to the origin of the 1918 'Spanish' influenza pandemic virus

Nonhomogeneous Markov models of nucleotide substitution have received scant attention. Here we explore the possibility of using nonhomogeneous models to identify host shift nodes along phylogenetic trees of pathogens evolving in different hosts. It has been noticed that influenza viruses show marked differences in nucleotide composition in human and avian hosts. We take advantage of this fact to identify the host shift event that led to the 1918 ‘Spanish’ influenza. This disease killed over 50 million people worldwide, ranking it as the deadliest pandemic in recorded history. Our model suggests that the eight RNA segments which eventually became the 1918 viral genome were introduced into a mammalian host around 1882–1913. The viruses later diverged into the classical swine and human H1N1 influenza lineages around 1913–1915. The last common ancestor of human strains dates from February 1917 to April 1918. Because pigs are more readily infected with avian influenza viruses than humans, it would seem that they were the original recipient of the virus. This would suggest that the virus was introduced into humans sometime between 1913 and 1918.

Panorama phylogenetic diversity and distribution of type A influenza viruses based on their six internal gene sequences

Background

Type A influenza viruses are important pathogens of humans, birds, pigs, horses and some marine mammals. The viruses have evolved into multiple complicated subtypes, lineages and sublineages. Recently, the phylogenetic diversity of type A influenza viruses from a whole view has been described based on the viral external HA and NA gene sequences, but remains unclear in terms of their six internal genes (PB2, PB1, PA, NP, MP and NS).

Methods

In this report, 2798 representative sequences of the six viral internal genes were selected from GenBank using the web servers in NCBI Influenza Virus Resource. Then, the phylogenetic relationships among the representative sequences were calculated using the software tools MEGA 4.1 and RAxML 7.0.4. Lineages and sublineages were classified mainly according to topology of the phylogenetic trees and distribution of the viruses in hosts, regions and time.

Results

The panorama phylogenetic trees of the six internal genes of type A influenza viruses were constructed. Lineages and sublineages within the type based on the six internal genes were classified and designated by a tentative universal numerical nomenclature system. The diversity of influenza viruses circulating in different regions, periods, and hosts based on the panorama trees was analyzed.

Conclusion

This study presents the first whole views to the phylogenetic diversity and distribution of type A influenza viruses based on their six internal genes. It also proposes a tentative universal nomenclature system for the viral lineages and sublineages. These can be a candidate framework to generalize the history and explore the future of the viruses, and will facilitate future scientific communications on the phylogenetic diversity and evolution of the viruses. In addition, it provides a novel phylogenetic view (i.e. the whole view) to recognize the viruses including the origin of the pandemic A(H1N1) influenza viruses.

Spatiotemporal dynamics in the early stages of the 2009 A/H1N1 influenza pandemic

Epidemiology and public health planning will increasingly rely on the analysis of genetic sequence data. The ongoing influenza A/H1N1 pandemic may represent a tipping point in this trend, with A/H1N1 being the first human pathogen routinely genotyped from the beginning of its spread. To take full advantage of this genetic information, we introduce a novel method to reconstruct the spatiotemporal dynamics of outbreaks from sequence data. The approach is based on a new paradigm were ancestries are inferred directly rather than through the reconstruction of most recent common ancestors (MRCAs) as in phylogenetics. Using 279 A/H1N1 hemagglutinin (HA) sequences, we confirm the emergence of the 2009 flu pandemic in Mexico. The virus initially spread to the US, and then to the rest of the world with both Mexico and the US acting as the main sources. While compatible with current epidemiological understanding of the 2009 H1N1 pandemic, our results provide a much finer picture of the spatiotemporal dynamics. The results also highlight how much additional epidemiological information can be gathered from genetic monitoring of a disease outbreak.

Influenza epidemiology and characterization of influenza viruses in patients seeking treatment for acute fever in Cambodia

The epidemiology, symptomology, and viral aetiology of endemic influenza remain largely uncharacterized in Cambodia. In December 2006, we established passive hospital-based surveillance to identify the causes of acute undifferentiated fever in patients seeking healthcare. Fever was defined as tympanic membrane temperature >38°C. From December 2006 to December 2008, 4233 patients were screened for influenza virus by real-time reverse-transcriptase polymerase chain reaction (rRT–PCR). Of these patients, 1151 (27·2%) were positive for influenza. Cough (68·8%vs. 50·5%,P<0·0001) and sore throat (55·0%vs. 41·9%,P<0·0001) were more often associated with laboratory-confirmed influenza-infected patients compared to influenza-negative enrollees. A clear influenza season was evident between July and December with a peak during the rainy season. Influenza A and B viruses were identified in 768 (66·3%) and 388 (33·7%) of the influenza-positive population (n=1153), respectively. In December 2008, passive surveillance identified infection of the avian influenza virus H5N1 in a 19-year-old farmer from Kandal province who subsequently recovered. From a subset of diagnostic samples submitted in 2007, 15 A(H1N1), seven A(H3N2) and seven B viruses were isolated. The predominant subtype tested was influenza A(H1N1), with the majority antigenically related to the A/Solomon Island/03/2006 vaccine strain. The influenza A(H3N2) isolates and influenza B viruses analysed were closely related to A/Brisbane/10/2007 or B/Ohio/01/2005 (B/Victoria/2/87-lineage) vaccine strains, respectively. Phylogenetic analysis of the HA1 region of the HA gene of influenza A(H1N1) viruses demonstrated that the Cambodian isolates belonged to clade 2C along with representative H1N1 viruses circulating in SE Asia at the time. These viruses remained sensitive to oseltamivir. In total, our data suggest that viral influenza infections contribute to nearly one-fifth of acute febrile illnesses and demonstrate the importance of influenza surveillance in Cambodia.

Directionality in the evolution of influenza A haemagglutinin

The evolution of haemagglutinin (HA), an important influenza virus antigen, has been the subject of intensive research for more than two decades. Many characteristics of HA's sequence evolution are captured by standard Markov chain substitution models. Such models assign equal fitness to all accessible amino acids at a site. We show, however, that such models strongly underestimate the number of homoplastic amino acid substitutions during the course of HA's evolution, i.e. substitutions that repeatedly give rise to the same amino acid at a site. We develop statistics to detect individual homoplastic events and find that they preferentially occur at positively selected epitopic sites. Our results suggest that the evolution of the influenza A HA, including evolution by positive selection, is strongly affected by the long-term site-specific preferences for individual amino acids.

Effects of single-point amino acid substitutions on the structure and function neuraminidase proteins in influenza A virus

In order to clarify the effect of amino acid substitutions on the structure and function of the neuraminidase (NA) protein of influenza A virus, we introduced single-point amino acid substitutions into the NA protein of the A/Tokyo/3/67 (H2N2) strain using PCR-based random mutation. The rate of tolerant random one amino acid substitutions in the NA protein was 47%. Rates of tolerant substitutions for the stalk and for the surface and inner portion of the head region of the NA protein were 79, 54, and 19%, respectively. Deleterious changes, such as those causing the NA protein to stop at the Golgi/endoplasmic reticulum, were scattered throughout the protein. On the other hand, the ratio of mutations with which the NA protein lost neuraminidase activity, but was transported to the cell surface, decreased in proportion to the distance from the structural center of enzyme active site. In order to investigate the effect of accumulated amino acid substitutions on the structural character of the N2NA protein during evolution, the same amino acid substitutions were introduced by site-directed mutagenesis at 23 homologous positions on N2 proteins of A/Tokyo/3/67, A/Bangkok/15/85 (H3N2), and A/Mie/1/2004 (H3N2). The results showed a shift, or discordance, in tolerance at some of the positions. An increase in discordance was correlated with the interval in years between virus strains, and the discordance rate was estimated to be 0.6-0.7% per year.

Exploring Variation in the d N /d S Ratio Among Sites and Lineages Using Mutational Mappings: Applications to the Influenza Virus

We use a likelihood-based method for mapping mutations on a phylogeny in a way that allows for both site-specific and lineage-specific variation in selection intensity. The method accounts for many of the potential sources of bias encountered in mapping of mutations on trees while still being computationally efficient. We apply the method to a previously published influenza data set to investigate hypotheses about changes in selection intensity in influenza strains. Influenza virus is sometimes propagated in chicken cells for several generations before sequencing, a process that has been hypothesized to induce mutations adapting the virus to the lab medium. Our analysis suggests that there are approximately twice as many replacement substitutions in lineages propagated in chicken eggs as in lineages that are not. Previous studies have attempted to predict which viral strains future epidemics may arise from using inferences regarding positive selection. The assumption is that future epidemics are more likely to arise from the strains in which positive selection on the so-called "trunk lineages" of the evolutionary tree is most pervasive. However, we find no difference in the strength of selection in the trunk lineages versus other evolutionary lineages. Our results suggest that it may be more difficult to use inferences regarding the strength of selection on mutations to make predictions regarding viral epidemics than previously thought.

Evolution and variation of the H3 gene of influenza A virus and interaction among hosts

OBJECTIVE

Continual mutations to the hemagglutinin (HA) gene of influenza A virus generate novel antigenic strains that cause annual epidemics. The aim of this study was to evaluate evolution tendency of the H3 gene in a long period of time.

METHODS

1842 H3 HA1 nucleotide strains of different hosts were collected for analysis. A two-step clustering method was used to divide strains into groups, and then a phylogenetic tree was constructed based on cluster results. Evolution rate in lineages were future estimated.

RESULTS

Tree structure showed three lineages: horse/canine, human/swine and avian. As a single trunk, the human/swine lineage was mainly composed of human strains, and more big branches appeared in recent years. Tree topology showed no evidence that swine affected the main evolution tendency of human H3 strains. The evolution rate of H3 strains varied between lineages. We observed that the rate in the human lineage decreased from 3.2 substitutions/year before 1980 to 1.8 after 1997.

CONCLUSION

We concluded that the variation of human H3 gene was associated with swine strains but independent of others, including bird strains. The evolution rate of human H3 strains seems to have decreased in recent years, while the reasons for the rate change need to be further explored.

[Accumulation of amino acid substitutions promotes irreversible structural changes in the hemagglutinin of human influenza AH3 virus during evolution]

During protein evolution the amino acid substitutions accumulate with time. However, the effect of accumulation of the amino acid substitutions to structural changes has not been estimated well. We will propose that the discordance of amino acid substitution on the HA protein of influenza A virus is useful for the assessment of structural changes during evolution. Discordance value can be obtained from the experimental data of tolerance or intolerance by introducing site directed mutagenesis at the homologous positions of two HA proteins holding the same amino acid residues. The value of discordance correlated to the number of amino acid differences among proteins. In the H3HA discordance rate was calculated to be 0.45% per one amino acid change. Furthermore, discordance of amino acid substitutions suggests that tolerable amino acid substitutions in different order have a probability of promoting irreversible divergence of the HA protein to different subtypes.

Natural Selection on the Influenza Virus Genome

Influenza viruses are the etiological agents of influenza. Although vaccines and drugs are available for the prophylaxis and treatment of influenza virus infections, the generation of escape mutants has been reported. To develop vaccines and drugs that are less susceptible to the generation of escape mutants, it is important to understand the evolutionary mechanisms of the viruses. Here natural selection operating on all the proteins encoded by the H3N2 human influenza A virus genome was inferred by comparing the numbers of synonymous (d(S) [D(S)]) and nonsynonymous (d(N) [D(N)]) substitutions per site. Natural selection was also inferred for the groups of functional amino acid sites involved in B-cell epitopes (BCEs), T-cell epitopes (TCEs), drug resistance, and growth in eggs. The entire region of PB1-F2 was positively selected, and positive selection also appeared to operate on BCEs, TCEs, and growth in eggs. The frequency of escape mutant generation appeared to be positively correlated with the d(N)/d(S) (D(N)/D(S)) values for the targets of vaccines and drugs, suggesting that the amino acid sites under strong functional constraint are suitable targets. In particular, TCEs may represent candidate targets because the d(N)/d(S) (D(N)/D(S)) values were small and negative selection was inferred for many of them.

Detailed analysis of the genetic evolution of influenza virus during the course of an epidemic

The genetic variability of influenza virus is usually studied with sequences selected over numerous years and countries, and rarely within a single season. Here we examined the viral evolution and the correlation between genetic and clinical features during an epidemic. From a French prospective household-based study in 1999-2000, 99 infected patients were randomly selected. The HA1 genomic domain was sequenced. Phylogenetic analysis showed the existence of two groups of A/H3N2 viruses. We found no distinct pattern of genomic evolution within either group according to time. A spatial correlation with the nucleotide distances was shown. The average nucleotide diversity was 3.4x10-3 nucleotides per site, and did not differ between the groups. A lower number of segregating sites was observed in patients who experienced influenza-like symptoms during the previous epidemic. These results suggest that the influenza virus undergoes regular HA1 nucleotide changes, but without clonal expansion of mutant strains within a single epidemic.

Isolation of virus from a SARS patient and genome-wide analysis of genetic mutations related to pathogenesis and epidemiology from 47 SARS-CoV isolates

Severe acute respiratory syndrome (SARS) caused by SARS-associated coronavirus (SARS-CoV) is a fatal disease. Prevention of future outbreaks is essential and requires understanding pathogenesis and evolution of the virus. We have isolated a SARS-CoV in China and analyzed 47 SARS-CoV genomes with the aims to reveal the evolution trends of the virus and provide insights into understanding pathogenesis and SARS epidemic. Specimen from a SARS patient was inoculated into cell culture. The presence of SARS-CoV was determined by RT-PCR and confirmed by electron microscopy. Virus was isolated followed by the determination of its genome sequences, which were then analyzed by comparing with other 46 SARS-CoV genomes. Genetic mutations with potential implications to pathogenesis and the epidemic were characterized. This viral genome consists of 29,728 nucleotides with overall organization in agreement with that of published isolates. A total of 348 positions were mutated on 47 viral genomes. Among them 22 had mutations in more than three genomes. Hot spots of nucleotide variations and unique trends of mutations were identified on the viral genomes. Mutation rates were different from gene to gene and were correlated well with periodical or geographic characteristics of the epidemic.

SARS-CoV infection was from at least two origins in the Taiwan area

Objective

Severe acute respiratory syndrome (SARS) is caused by a new coronavirus. Genomic sequence analysis will provide the molecular epidemiology and help to develop vaccines.

Methods

We developed a rapid method to amplify and sequence the whole SARS-CoV genome from clinical specimens. The technique employed one-step multiplex RT-PCR to amplify the whole SARS-CoV genome, and then nested PCR was performed to amplify a 2-kb region separately. The PCR products were sequenced.

Results

We sequenced the genomes of SARS-CoV from 3 clinical specimens obtained in Taiwan. The sequences were similar to those reported by other groups, except that 17 single nucleotide variations and two 2-nucleotide deletions, and a 1-nucleotide deletion were found. All the variations in the clinical specimens did not alter the amino acid sequence. Of these 17 sequenced variants, two loci (positions 26203 and 27812) were segregated together as a specific genotype-T:T or C:C. Phylogenetic analysis showed two major clusters of SARS patients in Taiwan.

conclusion

We developed a very economical and rapid method to sequence the whole genome of SARS-CoV, which can avoid cultural influence. From our results, SARS patients in Taiwan may be infected from two different origins.

Accumulation of amino acid substitutions promotes irreversible structural changes in the hemagglutinin of human influenza AH3 virus during evolution

In order to clarify the effect of an accumulation of amino acid substitutions on the hemadsorption character of the influenza AH3 virus hemagglutinin (HA) protein, we introduced single-point amino acid changes into the HA1 domain of the HA proteins of influenza viruses isolated in 1968 (A/Aichi/2/68) and 1997 (A/Sydney/5/97) by using PCR-based random mutation or site-directed mutagenesis. These substitutions were classified as positive or negative according to their effects on the hemadsorption activity. The rate of positive substitutions was about 50% for both strains. Of 44 amino acid changes that were identical in the two strains with regard to both the substituted amino acids and their positions in the HA1 domain, 22% of the changes that were positive in A/Aichi/2/68 were negative in A/Sydney/5/97 and 27% of the changes that were negative in A/Aichi/2/68 were positive in A/Sydney/5/97. A similar discordance rate was also seen for the antigenic sites. These results suggest that the accumulation of amino acid substitutions in the HA protein during evolution promoted irreversible structural changes and therefore that antigenic changes in the H3HA protein may not be limited.

Influenza as a model system for studying the cross-species transfer and evolution of the SARS coronavirus

Severe acute respiratory syndrome coronavirus (SARS-CoV) moved into humans from a reservoir species and subsequently caused an epidemic in its new host. We know little about the processes that allowed the cross-species transfer of this previously unknown virus. I discuss what we have learned about the movement of viruses into humans from studies of influenza A, both how it crossed from birds to humans and how it subsequently evolved within the human population. Starting with a brief review of severe acute respiratory syndrome to highlight the kinds of problems we face in learning about this viral disease, I then turn to influenza A, focusing on three topics. First, I present a reanalysis of data used to test the hypothesis that swine served as a "mixing vessel" or intermediate host in the transmission of avian influenza to humans during the 1918 "Spanish flu" pandemic. Second, I review studies of archived viruses from the three recent influenza pandemics. Third, I discuss current limitations in using molecular data to study the evolution of infectious disease. Although influenza A and SARS-CoV differ in many ways, our knowledge of influenza A may provide important clues about what limits or favours cross-species transfers and subsequent epidemics of newly emerging pathogens.

Three-dimensional window analysis for detecting positive selection at structural regions of proteins

Detection of natural selection operating at the amino acid sequence level is important in the study of molecular evolution. Single-site analysis and one-dimensional window analysis can be used to detect selection when the biological functions of amino acid sites are unknown. Single-site analysis is useful when selection operates more or less constantly over evolutionary time, but less so when selection operates temporarily. One-dimensional window analysis is more sensitive than single-site analysis when the functions of amino acid sites in close proximity in the linear sequence are similar, although this is not always the case. Here I present a three-dimensional window analysis method for detecting selection given the three-dimensional structure of the protein of interest. In the three-dimensional structure, the window is defined as the sphere centered on the alpha-carbon of an amino acid site. The window size is the radius of the sphere. The sites whose alpha-carbons are included in the window are grouped for the neutrality test. The window is moved within the three-dimensional structure by sequentially moving the central site along the primary amino acid sequence. To detect positive selection, it may also be useful to group the surface-exposed sites in the window separately. Three-dimensional window analysis appears not only to be more sensitive than single-site analysis and one-dimensional window analysis but also to provide similar specificity for inferring positive selection in the analyses of the hemagglutinin and neuraminidase genes of human influenza A viruses. This method, however, may fail to detect selection when it operates only on a particular site, in which case single-site analysis may be preferred, although a large number of sequences is required.

Moderate mutation rate in the SARS coronavirus genome and its implications

Background

The outbreak of severe acute respiratory syndrome (SARS) caused a severe global epidemic in 2003 which led to hundreds of deaths and many thousands of hospitalizations. The virus causing SARS was identified as a novel coronavirus (SARS-CoV) and multiple genomic sequences have been revealed since mid-April, 2003. After a quiet summer and fall in 2003, the newly emerged SARS cases in Asia, particularly the latest cases in China, are reinforcing a wide-spread belief that the SARS epidemic would strike back. With the understanding that SARS-CoV might be with humans for years to come, knowledge of the evolutionary mechanism of the SARS-CoV, including its mutation rate and emergence time, is fundamental to battle this deadly pathogen. To date, the speed at which the deadly virus evolved in nature and the elapsed time before it was transmitted to humans remains poorly understood.

Results

Sixteen complete genomic sequences with available clinical histories during the SARS outbreak were analyzed. After careful examination of multiple-sequence alignment, 114 single nucleotide variations were identified. To minimize the effects of sequencing errors and additional mutations during the cell culture, three strategies were applied to estimate the mutation rate by 1) using the closely related sequences as background controls; 2) adjusting the divergence time for cell culture; or 3) using the common variants only. The mutation rate in the SARS-CoV genome was estimated to be 0.80 – 2.38 × 10^-3 nucleotide substitution per site per year which is in the same order of magnitude as other RNA viruses. The non-synonymous and synonymous substitution rates were estimated to be 1.16 – 3.30 × 10^-3 and 1.67 – 4.67 × 10^-3 per site per year, respectively. The most recent common ancestor of the 16 sequences was inferred to be present as early as the spring of 2002.

Conclusions

The estimated mutation rates in the SARS-CoV using multiple strategies were not unusual among coronaviruses and moderate compared to those in other RNA viruses. All estimates of mutation rates led to the inference that the SARS-CoV could have been with humans in the spring of 2002 without causing a severe epidemic.