Search

Abstract Vaccines are the main pharmaceutical intervention used against the global public health threat posed by influenza viruses. Timely selection of optimal seed viruses with matched antigenicity between vaccine antigen and circulating viruses and with high yield underscore vaccine efficacy and supply, respectively. Current methods for selecting influenza seed vaccines are labor intensive and time-consuming. Here, we report the Machine-learning Assisted Influenza VaccinE Strain Selection framework, MAIVeSS, that enables streamlined selection of naturally circulating, antigenically matched, and high-yield influenza vaccine strains directly from clinical samples by using molecular signatures of antigenicity and yield to support optimal candidate vaccine virus selection. We apply our framework on publicly available sequences to select A(H1N1)pdm09 vaccine candidates and experimentally confirm that these candidates have optimal antigenicity and growth in cells and eggs. Our framework can potentially reduce the optimal vaccine candidate selection time from months to days and thus facilitate timely supply of seasonal vaccines.

Abstract SARS-CoV-2 is a zoonotic virus with documented bi-directional transmission between people and animals. Transmission of SARS-CoV-2 from humans to free-ranging white-tailed deer (Odocoileus virginianus) poses a unique public health risk due to the potential for reservoir establishment where variants may persist and evolve. We collected 8,830 respiratory samples from free-ranging white-tailed deer across Washington, D.C. and 26 states in the United States between November 2021 and April 2022. We obtained 391 sequences and identified 34 Pango lineages including the Alpha, Gamma, Delta, and Omicron variants. Evolutionary analyses showed these white-tailed deer viruses originated from at least 109 independent spillovers from humans, which resulted in 39 cases of subsequent local deer-to-deer transmission and three cases of potential spillover from white-tailed deer back to humans. Viruses repeatedly adapted to white-tailed deer with recurring amino acid substitutions across spike and other proteins. Overall, our findings suggest that multiple SARS-CoV-2 lineages were introduced, became enzootic, and co-circulated in white-tailed deer.

Abstract Escherichia coli (E. coli) are typically present as commensal bacteria in the gastro-intestinal tract of most animals including poultry species, but some avian pathogenic E. coli (APEC) strains can cause localized and even systematic infections in domestic poultry. Emergence and re-emergence of antimicrobial resistant isolates (AMR) constrain antibiotics usage in poultry production, and development of an effective vaccination program remains one of the primary options in E. coli disease prevention and control for domestic poultry. Thus, understanding genetic and pathogenic diversity of the enzootic E. coli isolates, particularly APEC, in poultry farms is the key to designing an optimal vaccine candidate and to developing an effective vaccination program. This study explored the genomic and pathogenic diversity among E. coli isolates in southern United States poultry. A total of nine isolates were recovered from sick broilers from Mississippi, and one from Georgia, with epidemiological variations among clinical signs, type of housing, and bird age. The genomes of these isolates were sequenced by using both Illumina short-reads and Oxford Nanopore long-reads, and our comparative analyses suggested data from both platforms were highly consistent. The 16 s rRNA based phylogenetic analyses showed that the 10 bacteria strains are genetically closer to each other than those in the public database. However, whole genome analyses showed that these 10 isolates encoded a diverse set of reported virulence and AMR genes, belonging to at least nine O:H serotypes, and are genetically clustered with at least five different groups of E. coli isolates reported by other states in the United States. Despite the small sample size, this study suggested that there was a large extent of genomic and serological diversity among E. coli isolates in southern United States poultry. A large-scale comprehensive study is needed to understand the overall genomic diversity and the associated virulence, and such a study will be important to develop a broadly protective E. coli vaccine.

SARS-CoV-2 has caused over 6.9 million deaths and continues to produce lasting health consequences. COVID-19 manifests broadly from no symptoms to death. In a retrospective cross-sectional study, we developed personalized risk assessment models that predict clinical outcomes for individuals with COVID-19 and inform targeted interventions. We sequenced viruses from SARS-CoV-2-positive nasopharyngeal swab samples between July 2020 and July 2022 from 4450 individuals in Missouri and retrieved associated disease courses, clinical history, and urban-rural classification. We integrated this data to develop machine learning-based predictive models to predict hospitalization, ICU admission, and long COVID.The mean age was 38.3 years (standard deviation = 21.4) with 55.2% (N = 2453) females and 44.8% (N = 1994) males (not reported, N = 4). Our analyses revealed a comprehensive set of predictors for each outcome, encompassing human, environment, and virus genome-wide genetic markers. Immunosuppression, cardiovascular disease, older age, cardiac, gastrointestinal, and constitutional symptoms, rural residence, and specific amino acid substitutions were associated with hospitalization. ICU admission was associated with acute respiratory distress syndrome, ventilation, bacterial co-infection, rural residence, and non-wild type SARS-CoV-2 variants. Finally, long COVID was associated with hospital admission, ventilation, and female sex.Overall, we developed risk assessment models that offer the capability to identify patients with COVID-19 necessitating enhanced monitoring or early interventions. Of importance, we demonstrate the value of including key elements of virus, host, and environmental factors to predict patient outcomes, serving as a valuable platform in the field of personalized medicine with the potential for adaptation to other infectious diseases.

ABSTRACT Millions of Norway rats (Rattus norvegicus) inhabit New York City (NYC), presenting the potential for transmission of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) from humans to rats. We evaluated SARS-CoV-2 exposure among 79 rats captured from NYC during the fall of 2021. Our results showed that 13 of the 79 rats (16.5%) tested IgG- or IgM-positive, and partial SARS-CoV-2 genomes were recovered from all 4 rats that were qRT-PCR (reverse transcription-quantitative PCR)-positive. Genomic analyses suggest these viruses were associated with genetic lineage B, which was predominant in NYC in the spring of 2020 during the early pandemic period. To further investigate rat susceptibility to SARS-CoV-2 variants, we conducted a virus challenge study and showed that Alpha, Delta, and Omicron variants can cause infections in wild-type Sprague Dawley (SD) rats, including high replication levels in the upper and lower respiratory tracts and induction of both innate and adaptive immune responses. Additionally, the Delta variant resulted in the highest infectivity. In summary, our results indicate that rats are susceptible to infection with Alpha, Delta, and Omicron variants, and wild Norway rats in the NYC municipal sewer systems have been exposed to SARS-CoV-2. Our findings highlight the need for further monitoring of SARS-CoV-2 in urban rat populations and for evaluating the potential risk of secondary zoonotic transmission from these rat populations back to humans. IMPORTANCE The host tropism expansion of SARS-CoV-2 raises concern for the potential risk of reverse-zoonotic transmission of emerging variants into rodent species, including wild rat species. In this study, we present both genetic and serological evidence for SARS-CoV-2 exposure to the New York City wild rat population, and these viruses may be linked to the viruses that were circulating during the early stages of the pandemic. We also demonstrated that rats are susceptible to additional variants (i.e., Alpha, Delta, and Omicron) that have been predominant in humans and that susceptibility to infection varies by variant. Our findings highlight the reverse zoonosis of SARS-CoV-2 to urban rats and the need for further monitoring of SARS-CoV-2 in rat populations for potential secondary zoonotic transmission to humans.

Abstract Background In addition to causing the pandemic influenza outbreaks of 1918 and 2009, subtype H1N1 influenza A viruses (IAVs) have caused seasonal epidemics since 1977. Antigenic property of influenza viruses are determined by both protein sequence and N-linked glycosylation of influenza glycoproteins, especially hemagglutinin (HA). The currently available computational methods are only considered features in protein sequence but not N-linked glycosylation. Results A multi-task learning sparse group least absolute shrinkage and selection operator (LASSO) (MTL-SGL) regression method was developed and applied to derive two types of predominant features including protein sequence and N-linked glycosylation in hemagglutinin (HA) affecting variations in serologic data for human and swine H1N1 IAVs. Results suggested that mutations and changes in N-linked glycosylation sites are associated with the rise of antigenic variants of H1N1 IAVs. Furthermore, the implicated mutations are predominantly located at five reported antibody-binding sites, and within or close to the HA receptor binding site. All of the three N-linked glycosylation sites (i.e. sequons NCSV at HA 54, NHTV at HA 125, and NLSK at HA 160) identified by MTL-SGL to determine antigenic changes were experimentally validated in the H1N1 antigenic variants using mass spectrometry analyses. Compared with conventional sparse learning methods, MTL-SGL achieved a lower prediction error and higher accuracy, indicating that grouped features and MTL in the MTL-SGL method are not only able to handle serologic data generated from multiple reagents, supplies, and protocols, but also perform better in genetic sequence-based antigenic quantification. Conclusions In summary, the results of this study suggest that mutations and variations in N-glycosylation in HA caused antigenic variations in H1N1 IAVs and that the sequence-based antigenicity predictive model will be useful in understanding antigenic evolution of IAVs.

Compatibility among the influenza A virus (IAV) ribonucleoprotein (RNP) genes affects viral replication efficiency and can limit the emergence of novel reassortants, including those with potential pandemic risks. In this study, we determined the polymerase activities of 2,451 RNP reassortants among three seasonal and eight enzootic IAVs by using a minigenome assay. Results showed that the 2009 H1N1 RNP are more compatible with the tested enzootic RNP than seasonal H3N2 RNP and that triple reassortment increased such compatibility. The RNP reassortants among 2009 H1N1, canine H3N8, and avian H4N6 IAVs had the highest polymerase activities. Residues in the RNA binding motifs and the contact regions among RNP proteins affected polymerase activities. Our data indicates that compatibility among seasonal and enzootic RNPs are selective, and enzoosis of multiple strains in the animal-human interface can facilitate emergence of an RNP with increased replication efficiency in mammals, including humans.

Subtype H6 avian influenza A viruses (IAVs) are enzootic and genetically diverse in both domestic poultry and wild waterfowl and may cause spillovers in both pigs and humans. Thus, it is important to understand the genetic diversity of H6 IAVs in birds and their zoonotic potential. Compared with that in domestic poultry, the genetic diversity of H6 viruses in wild birds in China has not been well-understood. In this study, five H6 viruses were isolated from wild birds in Poyang Lake, China, and genetic analyses showed that these isolates are clustered into four genotypes associated with reassortments among avian IAVs from domestic poultry and wild birds in China and those from Eurasia and North America and that these viruses exhibited distinct phenotypes in growth kinetics analyses with avian and mammalian cells lines and in mouse challenge experiments. Of interest is that two H6 isolates from the Eurasian teal replicated effectively in the mouse lung without prior adaptation, whereas the other three did not. Our study suggested that there are variations in the mammalian viral replication efficiency phenotypic among genetically diverse H6 IAVs in wild birds and that both intra- and inter-continental movements of IAVs through wild bird migration may facilitate the emergence of novel H6 IAV reassortants with the potential for replicating in mammals, including humans. Continued surveillance to monitor the diversity of H6 IAVs in wild birds is necessary to increase our understanding of the natural history of IAVs.

Human plasma fibronectin is an adhesive protein that plays a crucial role in wound healing. Many studies had indicated that glycans might mediate the expression and functions of fibronectin, yet a comprehensive understanding of its glycosylation is still missing. Here, we performed a comprehensive N- and O-glycosylation mapping of human plasma fibronectin and quantified the occurrence of each glycoform in a site-specific manner. Intact N-glycopeptides were enriched by zwitterionic hydrophilic interaction chromatography, and N-glycosite sites were localized by the 18O-labeling method. O-glycopeptide enrichment and O-glycosite identification were achieved by an enzyme-assisted site-specific extraction method. An RP–LC–MS/MS system functionalized with collision-induced dissociation and stepped normalized collision energy (sNCE)-HCD tandem mass was applied to analyze the glycoforms of fibronectin. A total of 6 N-glycosites and 53 O-glycosites were identified, which were occupied by 38 N-glycoforms and 16 O-glycoforms, respectively. Furthermore, 77.31% of N-glycans were sialylated, and O-glycosylation was dominated by the sialyl-T antigen. These site-specific glycosylation patterns on human fibronectin can facilitate functional analyses of fibronectin and therapeutics development.

[This corrects the article DOI: 10.1371/journal.ppat.1008421.].

In humans and other mammals, the respiratory tract is represented by a complex network of polarized epithelial cells, forming an apical surface facing the external environment and a basal surface attached to the basement layer. These cells are characterized by differential expression of proteins and glycans, which serve as receptors during influenza virus infection. Attachment between these host receptors and the viral surface glycoprotein hemagglutinin (HA) initiates the influenza virus life cycle. However, the virus receptor binding specificities may not be static. Sialylated N-glycans are the most well-characterized receptors but are not essential for the entry of influenza viruses, and other molecules, such as O-glycans and non-sialylated glycans, may be involved in virus-cell attachment. Furthermore, correct cell polarity and directional trafficking of molecules are essential for the orderly development of the system and affect successful influenza infection; on the other hand, influenza infection can also change cell polarity. Here we review recent advances in our understanding of influenza virus infection in the respiratory tract of humans and other mammals, particularly the attachment between the virus and the surface of the polar cells and the polarity variation of these cells due to virus infection.

Siglecs are sialic acid–binding immunoglobulin-like lectins that play vital roles in immune cell signaling. Siglecs help the immune system distinguish between self and nonself through the recognition of glycan ligands. While the primary binding specificities of Siglecs are known to be divergent, their specificities for complex glycans remain unclear. Herein, we determined N-glycan binding profiles of a set of Siglecs by using a complex asymmetric N-glycan microarray. Our results showed that Siglecs had unique terminal epitope-dependent branch preference when recognizing asymmetric N-glycans. Specifically, human Siglec-3, -9, and -10 prefer the α1-3 branch when Siaα2-6Galβ1-4GlcNAc terminal epitope serves as the binding ligand but prefer the opposite α1-6 branch when Siaα2-3Galβ1-4GlcNAc epitope serves as the ligand. Interestingly, Siglec-10 exhibited dramatic binding divergence toward a pair of Neu5Ac-containing asymmetric N-glycan isomers, as well as their Neu5Gc-containing counterparts. This new information on complex glycan recognition by Siglecs provides insights into their biological roles and applications.

Both influenza A virus (IAV) and influenza D virus (IDV) are enzootic in pigs. IAV causes approximately 100% morbidity with low mortality, whereas IDV leads to only mild respiratory diseases in pigs. In this study, we performed a series of coinfection experiments in vitro and in vivo to understand how IAV and IDV interact and cause pathogenesis during coinfection. The results showed that IAV inhibited IDV replication when infecting swine tracheal epithelial cells (STECs) with IAV 24 or 48 h prior to IDV inoculation and that IDV suppressed IAV replication when IDV preceded IAV inoculation by 48 h. Virus interference was not identified during simultaneous IAV/IDV infections or with 6 h between the two viral infections, regardless of their order. The interference pattern at 24 and 48 h correlated with proinflammatory responses induced by the first infection, which, for IDV, was slower than for IAV by about 24 h. The viruses did not interfere with each other if both infected the cells before proinflammatory responses were induced. Coinfection in pigs further demonstrated that IAV interfered with both viral shedding and virus replication of IDV, especially in the upper respiratory tract. Clinically, coinfection of IDV and IAV did not show significant enhancement of disease pathogenesis, compared with the pigs infected with IAV alone. In summary, this study suggests that interference during coinfection of IAV and IDV is primarily due to the proinflammatory response; therefore, it is dependent on the time between infections and the order of infection. This study facilitates our understanding of virus epidemiology and pathogenesis associated with IAV and IDV coinfection.

Pandemic influenza A viruses can emerge from swine, an intermediate host that supports adaptation of human-preferred receptor-binding specificity by the hemagglutinin (HA) surface antigen. Other HA traits necessary for pandemic potential are poorly understood. For swine influenza viruses isolated in 2009–2016, gamma-clade viruses had less stable HA proteins (activation pH 5.5–5.9) than pandemic clade (pH 5.0–5.5). Gamma-clade viruses replicated to higher levels in mammalian cells than pandemic clade. In ferrets, a model for human adaptation, a relatively stable HA protein (pH 5.5–5.6) was necessary for efficient replication and airborne transmission. The overall airborne transmission frequency in ferrets for four isolates tested was 42%, and isolate G15 airborne transmitted 100% after selection of a variant with a stabilized HA. The results suggest swine influenza viruses containing both a stabilized HA and alpha-2,6 receptor binding in tandem pose greater pandemic risk. Increasing evidence supports adding HA stability to pre-pandemic risk assessment algorithms.

The outbreak of a novel corona Virus Disease 2019 (COVID-19) in the city of Wuhan, China has resulted in more than 1.7 million laboratory confirmed cases all over the world. Recent studies showed that SARS-CoV-2 was likely originated from bats, but its intermediate hosts are still largely unknown. In this study, we assembled the complete genome of a coronavirus identified in 3 sick Malayan pangolins. The molecular and phylogenetic analyses showed that this pangolin coronavirus (pangolin-CoV-2020) is genetically related to the SARS-CoV-2 as well as a group of bat coronaviruses but do not support the SARS-CoV-2 emerged directly from the pangolin-CoV-2020. Our study suggests that pangolins are natural hosts of Betacoronaviruses. Large surveillance of coronaviruses in pangolins could improve our understanding of the spectrum of coronaviruses in pangolins. In addition to conservation of wildlife, minimizing the exposures of humans to wildlife will be important to reduce the spillover risks of coronaviruses from wild animals to humans.

Influenza D virus (IDV) has been identified in domestic cattle, swine, camelid, and small ruminant populations across North America, Europe, Asia, South America, and Africa. Our study investigated seroprevalence and transmissibility of IDV in feral swine. During 2012–2013, we evaluated feral swine populations in 4 US states; of 256 swine tested, 57 (19.1%) were IDV seropositive. Among 96 archived influenza A virus–seropositive feral swine samples collected from 16 US states during 2010–2013, 41 (42.7%) were IDV seropositive. Infection studies demonstrated that IDV-inoculated feral swine shed virus 3–5 days postinoculation and seroconverted at 21 days postinoculation; 50% of in-contact naive feral swine shed virus, seroconverted, or both. Immunohistochemical staining showed viral antigen within epithelial cells of the respiratory tract, including trachea, soft palate, and lungs. Our findings suggest that feral swine might serve an important role in the ecology of IDV.

Influenza B viruses (IBVs) are causing an increasing burden of morbidity and mortality, yet the prevalence of culture-adapted mutations in human seasonal IBVs are unclear. We collected 368 clinical samples from patients with influenza-like illness in Missouri during the 2019–2020 influenza season and recovered 146 influenza isolates including 38 IBV isolates. Of MDCK-CCL34, MDCK-Siat1, and humanized MDCK (hCK), hCK showed the highest virus recovery efficiency. All Missourian IBVs belonged to the Victoria V1A.3 lineage, all of which contained a three-amino acid deletion on the HA protein and were antigenically distant from the Victoria lineage IBV vaccine strain used during that season. By comparing genomic sequences of these IBVs in 31 paired samples, eight cell-adapted nonsynonymous mutations were identified, with the majority in the RNA polymerase. Analyses of IBV clinical sample–isolate pairs from public databases further showed that cell- and egg-adapted mutations occurred more widely in viral proteins, including the receptor and antibody binding sites on HA. Our study suggests that hCK is an effective platform for IBV isolation and that culture-adapted mutations may occur during IBV isolation. As culture-adapted mutations may affect subsequent virus studies and vaccine development, the knowledge from this study may help optimize strategies for influenza surveillance, vaccine strain selection, and vaccine development.

Subtype H6 influenza A viruses (IAVs) are commonly detected in wild birds and domestic poultry and can infect humans. In 2010, a H6N6 virus emerged in southern China, and since then, it has caused sporadic infections among swine. We show that this virus binds to α2,6-linked and α2,3-linked sialic acids. Mutations at residues 222 (alanine to valine) and 228 (glycine to serine) of the virus hemagglutinin (HA) affected its receptor-binding properties. Experiments showed that the virus has limited transmissibility between ferrets through direct contact or through inhalation of infectious aerosolized droplets. The internal genes of the influenza A(H1N1)pdm09 virus, which is prevalent in swine worldwide, increases the replication efficiency of H6N6 IAV in the lower respiratory tract of ferrets but not its transmissibility between ferrets. These findings suggest H6N6 swine IAV (SIV) currently poses a moderate risk to public health, but its evolution and spread should be closely monitored.Emerging Microbes & Infections (2017) 6, e17; doi:10.1038/emi.2017.3; published online 12 April 2017

Genetic reassortment between influenza A viruses (IAVs) facilitate emergence of pandemic strains, and swine are proposed as a "mixing vessel" for generating reassortants of avian and mammalian IAVs that could be of risk to mammals, including humans. However, how a transmissible reassortant emerges in swine are not well understood. Genomic analyses of 571 isolates recovered from nasal wash samples and respiratory tract tissues of a group of co-housed pigs (influenza-seronegative, avian H1N1 IAV-infected, and swine H3N2 IAV-infected pigs) identified 30 distinct genotypes of reassortants. Viruses recovered from lower respiratory tract tissues had the largest genomic diversity, and those recovered from turbinates and nasal wash fluids had the least. Reassortants from lower respiratory tracts had the largest variations in growth kinetics in respiratory tract epithelial cells, and the cold temperature in swine nasal cells seemed to select the type of reassortant viruses shed by the pigs. One reassortant in nasal wash samples was consistently identified in upper, middle, and lower respiratory tract tissues, and it was confirmed to be transmitted efficiently between pigs. Study findings suggest that, during mixed infections of avian and swine IAVs, genetic reassortments are likely to occur in the lower respiratory track, and tissue tropism is an important factor selecting for a transmissible reassortant.

To optimize the public health response to coronavirus disease 2019 (COVID-19), we must first understand the antibody response to individual proteins on the severe acute respiratory syndrome-related coronavirus 2 (SARS-CoV-2) and the antibody’s cross reactivity to other coronaviruses. Using a panel of 37 convalescent COVID-19 human serum samples, we showed that the magnitude and specificity of responses varied across individuals, independent of their reactivity to seasonal human coronaviruses (HCoVs). These data suggest that COVID-19 vaccines will elicit primary humoral immune responses in naïve individuals and variable responses in those previously exposed to SARS-CoV-2. Unlike the limited cross-coronavirus reactivities in humans, serum samples from 96 dogs and 10 cats showed SARS-CoV-2 protein-specific responses focused on non–S1 proteins. The correlation of this response with those to other coronaviruses suggests that the antibodies are cross-reactive and generated to endemic viruses within these hosts, which must be considered in seroepidemiologic studies. We conclude that substantial variation in antibody generation against coronavirus proteins will influence interpretations of serologic data in the clinical and veterinary settings.

To determine whether, and to what extent, influenza A subtype H3 viruses were present in feral swine in the United States, we conducted serologic and virologic surveillance during October 2011–September 2012. These animals were periodically exposed to and infected with A(H3N2) viruses, suggesting they may threaten human and animal health.

Rapid identification of influenza antigenic variants will be critical in selecting optimal vaccine candidates and thus a key to developing an effective vaccination program. Recent studies suggest that multiple simultaneous mutations at antigenic sites accumulatively enhance antigenic drift of influenza A viruses. However, pre-existing methods on antigenic variant identification are based on analyses from individual sites. Because the impacts of these co-evolved sites on influenza antigenicity may not be additive, it will be critical to quantify the impact of not only those single mutations but also multiple simultaneous mutations or co-evolved sites. Here, we developed and applied a computational method, AntigenCO, to identify and quantify both single and co-evolutionary sites driving the historical antigenic drifts. AntigenCO achieved an accuracy of up to 90.05% for antigenic variant prediction, significantly outperforming methods based on single sites. AntigenCO can be useful in antigenic variant identification in influenza surveillance.

ABSTRACT The efficacy of current influenza vaccines requires a close antigenic match between circulating and vaccine strains. As such, timely identification of emerging influenza virus antigenic variants is central to the success of influenza vaccination programs. Empirical methods to determine influenza virus antigenic properties are time-consuming and mid-throughput and require live viruses. Here, we present a novel, experimentally validated, computational method for determining influenza virus antigenicity on the basis of hemagglutinin (HA) sequence. This method integrates a bootstrapped ridge regression with antigenic mapping to quantify antigenic distances by using influenza HA1 sequences. Our method was applied to H3N2 seasonal influenza viruses and identified the 13 previously recognized H3N2 antigenic clusters and the antigenic drift event of 2009 that led to a change of the H3N2 vaccine strain. IMPORTANCE This report supplies a novel method for quantifying antigenic distance and identifying antigenic variants using sequences alone. This method will be useful in influenza vaccine strain selection by significantly reducing the human labor efforts for serological characterization and will increase the likelihood of correct influenza vaccine candidate selection.

A striking characteristic of the past four influenza pandemic outbreaks in the United States has been the multiple waves of infections. However, the mechanisms responsible for the multiple waves of influenza or other acute infectious diseases are uncertain. Understanding these mechanisms could provide knowledge for health authorities to develop and implement prevention and control strategies.We exhibit five distinct mechanisms, each of which can generate two waves of infections for an acute infectious disease. The first two mechanisms capture changes in virus transmissibility and behavioral changes. The third mechanism involves population heterogeneity (e.g., demography, geography), where each wave spreads through one sub-population. The fourth mechanism is virus mutation which causes delayed susceptibility of individuals. The fifth mechanism is waning immunity. Each mechanism is incorporated into separate mathematical models, and outbreaks are then simulated. We use the models to examine the effects of the initial number of infected individuals (e.g., border control at the beginning of the outbreak) and the timing of and amount of available vaccinations.Four models, individually or in any combination, reproduce the two waves of the 2009 H1N1 pandemic in the United States, both qualitatively and quantitatively. One model reproduces the two waves only qualitatively. All models indicate that significantly reducing or delaying the initial numbers of infected individuals would have little impact on the attack rate. Instead, this reduction or delay results in a single wave as opposed to two waves. Furthermore, four of these models also indicate that a vaccination program started earlier than October 2009 (when the H1N1 vaccine was initially distributed) could have eliminated the second wave of infection, while more vaccine available starting in October would not have eliminated the second wave.

Antigenic characterization based on serological data, such as Hemagglutination Inhibition (HI) assay, is one of the routine procedures for influenza vaccine strain selection. In many cases, it would be impossible to measure all pairwise antigenic correlations between testing antigens and reference antisera in each individual experiment. Thus, we have to combine and integrate the HI tables from a number of individual experiments. Measurements from different experiments may be inconsistent due to different experimental conditions. Consequently we will observe a matrix with missing data and possibly inconsistent measurements. In this paper, we develop a new mathematical model, which we refer to as Joint Matrix Completion and Filtering, for HI data integration. In this approach, we simultaneously handle the incompleteness and uncertainty of observations by assuming that the underlying merged HI data matrix has low rank, as well as carefully modeling different levels of noises in each individual table. An efficient blockwise coordinate descent procedure is developed for optimization. The performance of our approach is validated on synthetic and real influenza datasets. The proposed joint matrix completion and filtering model can be adapted as a general model for biological data integration, targeting data noises and missing values within and across experiments.

The H1N1 subtype of influenza A virus has caused two of the four documented pandemics and is responsible for seasonal epidemic outbreaks, presenting a continuous threat to public health. Co-circulating antigenically divergent influenza strains significantly complicates vaccine development and use. Here, by combining evolutionary, structural, functional, and population information about the H1N1 proteome, we seek to answer two questions: (1) do residues on the protein surfaces evolve faster than the protein core residues consistently across all proteins that constitute the influenza proteome? and (2) in spite of the rapid evolution of surface residues in influenza proteins, are there any protein regions on the protein surface that do not evolve? To answer these questions, we first built phylogenetically-aware models of the patterns of surface and interior substitutions. Employing these models, we found a single coherent pattern of faster evolution on the protein surfaces that characterizes all influenza proteins. The pattern is consistent with the events of inter-species reassortment, the worldwide introduction of the flu vaccine in the early 80's, as well as the differences caused by the geographic origins of the virus. Next, we developed an automated computational pipeline to comprehensively detect regions of the protein surface residues that were 100% conserved over multiple years and in multiple host species. We identified conserved regions on the surface of 10 influenza proteins spread across all avian, swine, and human strains; with the exception of a small group of isolated strains that affected the conservation of three proteins. Surprisingly, these regions were also unaffected by genetic variation in the pandemic 2009 H1N1 viral population data obtained from deep sequencing experiments. Finally, the conserved regions were intrinsically related to the intra-viral macromolecular interaction interfaces. Our study may provide further insights towards the identification of novel protein targets for influenza antivirals.

Influenza viruses have been responsible for large losses of lives around the world and continue to present a great public health challenge. Antigenic characterization based on hemagglutination inhibition (HI) assay is one of the routine procedures for influenza vaccine strain selection. However, HI assay is only a crude experiment reflecting the antigenic correlations among testing antigens (viruses) and reference antisera (antibodies). Moreover, antigenic characterization is usually based on more than one HI dataset. The combination of multiple datasets results in an incomplete HI matrix with many unobserved entries. This paper proposes a new computational framework for constructing an influenza antigenic cartography from this incomplete matrix, which we refer to as Matrix Completion-Multidimensional Scaling (MC-MDS). In this approach, we first reconstruct the HI matrices with viruses and antibodies using low-rank matrix completion, and then generate the two-dimensional antigenic cartography using multidimensional scaling. Moreover, for influenza HI tables with herd immunity effect (such as those from Human influenza viruses), we propose a temporal model to reduce the inherent temporal bias of HI tables caused by herd immunity. By applying our method in HI datasets containing H3N2 influenza A viruses isolated from 1968 to 2003, we identified eleven clusters of antigenic variants, representing all major antigenic drift events in these 36 years. Our results showed that both the completed HI matrix and the antigenic cartography obtained via MC-MDS are useful in identifying influenza antigenic variants and thus can be used to facilitate influenza vaccine strain selection. The webserver is available at http://sysbio.cvm.msstate.edu/AntigenMap.

Vietnam is one of the countries most affected by outbreaks of H5N1 highly pathogenic avian influenza viruses. First identified in Vietnam in poultry in 2001 and in humans in 2004, the virus has since caused 111 cases and 56 deaths in humans. In 2003/2004 H5N1 outbreaks, nearly the entire poultry population of Vietnam was culled. Our earlier study (Wan et al., 2008, PLoS ONE, 3(10): e3462) demonstrated that there have been at least six independent H5N1 introductions into Vietnam and there were nine newly emerged reassortants from 2001 to 2007 in Vietnam. H5N1 viruses in Vietnam cluster distinctly around Hanoi and Ho Chi Minh City. However, the nature of the relationship between genetic divergence and geographic patterns is still unclear.In this study, we hypothesized that genetic distances between H5N1 viruses in Vietnam are correlated with geographic distances, as the result of distinct population and environment patterns along Vietnam's long north to south longitudinal extent. Based on this hypothesis, we combined spatial statistical methods with genetic analytic techniques and explicitly used geographic space to explore genetic evolution of H5N1 highly pathogenic avian influenza viruses at the sub-national scale in Vietnam. Our dataset consisted of 125 influenza viruses (with whole genome sets) isolated in Vietnam from 2003 to 2007. Our results document the significant effect of space and time on genetic evolution and the rise of two regional centers of genetic mixing by 2007. These findings give insight into processes underlying viral evolution and suggest that genetic differentiation is associated with the distance between concentrations of human and poultry populations around Hanoi and Ho Chi Minh City.The results show that genetic evolution of H5N1 viruses in Vietnamese domestic poultry is highly correlated with the location and spread of those viruses in geographic space. This correlation varies by scale, time, and gene, though a classic isolation by distance pattern is observed. This study is the first to characterize the geographic structure of influenza viral evolution at the sub-national scale in Vietnam and can shed light on how H5N1 HPAIVs evolve in certain geographic settings.

Highly pathogenic avian influenza (HPAI) H5N1 viruses have caused dramatic economic losses to the poultry industry of Vietnam and continue to pose a serious threat to public health. As of June 2008, Vietnam had reported nearly one third of worldwide laboratory confirmed human H5N1 infections. To better understand the emergence, spread and evolution of H5N1 in Vietnam we studied over 300 H5N1 avian influenza viruses isolated from Vietnam since their first detection in 2001. Our phylogenetic analyses indicated that six genetically distinct H5N1 viruses were introduced into Vietnam during the past seven years. The H5N1 lineage that evolved following the introduction in 2003 of the A/duck/Hong Kong/821/2002-like viruses, with clade 1 hemagglutinin (HA), continued to predominate in southern Vietnam as of May 2007. A virus with a clade 2.3.4 HA newly introduced into northern Vietnam in 2007, reassorted with pre-existing clade 1 viruses, resulting in the emergence of novel genotypes with neuraminidase (NA) and/or internal gene segments from clade 1 viruses. A total of nine distinct genotypes have been present in Vietnam since 2001, including five that were circulating in 2007. At least four of these genotypes appear to have originated in Vietnam and represent novel H5N1 viruses not reported elsewhere. Geographic and temporal analyses of H5N1 infection dynamics in poultry suggest that the majority of viruses containing new genes were first detected in northern Vietnam and subsequently spread to southern Vietnam after reassorting with pre-existing local viruses in northern Vietnam. Although the routes of entry and spread of H5N1 in Vietnam remain speculative, enhanced poultry import controls and virologic surveillance efforts may help curb the entry and spread of new HPAI viral genes.