Identifying antigenicity-associated sites in highly pathogenic H5N1 influenza virus hemagglutinin by using sparse learning

Genetic features of avian influenza (A/H5N8) clade 2.3.4.4b isolated from quail in Egypt

Several genotypes of the highly pathogenic avian influenza (HPAI) virus H5N8 subtype within clade 2.3.4.4b continue to circulate in different species of domestic birds across Egypt. It is believed that quail contribute to virus replication and adaptation to other gallinaceous poultry species and humans. This study provides genetic characterization of the full genome of HPAI H5N8 isolated from quail in Egypt. The virus was isolated from a commercial quail farm associated with respiratory signs. To characterize the genetic features of the detected virus, gene sequencing via Sanger technology and phylogenetic analysis were performed. The results revealed high nucleotide identity with the HPAI H5N8 virus from Egypt, which has multiple basic amino acid motifs PLREKRRKR/GLF at the hemagglutinin (HA) cleavage site. Phylogenetic analysis of the eight gene segments revealed that the quail isolate is grouped with HPAI H5N8 viruses of clade 2.3.4.4b and closely related to the most recent circulating H5N8 viruses in Egypt. Whole-genome characterization revealed amino acid preferences for avian receptors with few mutations, indicating their affinity for human-like receptors and increased virulence in mammals, such as S123P, S133A, T156A and A263T in the HA gene. In addition, the sequencing results revealed a lack of markers associated with influenza antiviral resistance in the neuraminidase and matrix-2 coding proteins. The results of the present study support the spread of HPAIV H5N8 to species other than chickens in Egypt. Therefore, continuous surveillance of AIV in different bird species in Egypt followed by full genomic characterization is needed for better virus control and prevention.

Continuing evolution of H5N1 highly pathogenic avian influenza viruses of clade 2.3.2.1a G2 genotype in domestic poultry of Bangladesh during 2018-2021

We characterized 15 H5N1 HPAI viruses from different small- and medium-scale poultry flocks across Bangladesh during 2018-2021 based on their complete genome sequences. The antigenic relatedness of H5N1 HPAI viruses from different timepoints was analysed. During 2020-2021, 42.11% of the flocks tested positive for at least one of the respiratory infections, with 15.79% showing influenza A virus, of which 8.77% tested positive for HPAIV H5N1. Co-infections with two to four pathogens were detected in 15.8% of flocks. Phylogeny and gene constellation analyses based on complete genome sequences of 15 HPAI viruses revealed the continuing circulation of H5 clade 2.3.2.1a genotype G2 viruses. In the HA protein of the study isolates, functionally meaningful mutations caused the loss of an N-linked glycosylation site (T156A), a modified antigenic site A (S141P), and a mutation in the receptor binding pocket (E193R/K). Consequently, antigenic analysis revealed a significant loss of cross-reactivity between viruses from different host species and periods. Most viruses displayed oseltamivir resistance markers at positions V96, I97, S227, and N275 (N1 numbering) of the NA protein. In addition, for the PB2, M1, and NS1 proteins, significant mutations were noticed that have been associated with polymerase activity and increased virulence for mammals in all study isolates. These results highlight the need for intensified genomic surveillance of HPAI circulating in poultry in Bangladesh and for establishing appropriate control measures to decrease the circulation of these viruses in poultry in the country.

N-glycosylation on hemagglutinin head reveals inter-branch antigenic variability of avian influenza virus H5-subtypes

H5-subtype avian influenza virus (AIV) is globally prevalent and undergoes frequent antigenic drift, necessitating regular updates to vaccines. One of the many influencing elements that cause incompatibility between vaccinations and epidemic strains is the dynamic alteration of glycosylation sites. However, the biological significance of N-glycosylation in the viral evolution and antigenic changes is unclear. Here, we performed a systematic analysis of glycosylation sites on the HA1 subunit of H5N1, providing insights into the changes of primary glycosylation sites, including 140 N, 156 N, and 170 N within the antigenic epitopes of HA1 protein. Multiple recombinant viruses were then generated based on HA genes of historical vaccine strains and deactivated for immunizing SPF chickens. Inactivated recombinant strains showed relatively closer antigenicity compared to which has identical N-glycosylation patterns. The N-glycosylation modification discrepancy highlights the inter-branch antigenic diversity of H5-subtype viruses in avian influenza and serves as a vital foundation for improving vaccination tactics.

MAIVeSS: streamlined selection of antigenically matched, high-yield viruses for seasonal influenza vaccine production

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.

A sequence-based machine learning model for predicting antigenic distance for H3N2 influenza virus

Introduction

Seasonal influenza A H3N2 viruses are constantly changing, reducing the effectiveness of existing vaccines. As a result, the World Health Organization (WHO) needs to frequently update the vaccine strains to match the antigenicity of emerged H3N2 variants. Traditional assessments of antigenicity rely on serological methods, which are both labor-intensive and time-consuming. Although numerous computational models aim to simplify antigenicity determination, they either lack a robust quantitative linkage between antigenicity and viral sequences or focus restrictively on selected features.

Methods

Here, we propose a novel computational method to predict antigenic distances using multiple features, including not only viral sequence attributes but also integrating four distinct categories of features that significantly affect viral antigenicity in sequences.

Results

This method exhibits low error in virus antigenicity prediction and achieves superior accuracy in discerning antigenic drift. Utilizing this method, we investigated the evolution process of the H3N2 influenza viruses and identified a total of 21 major antigenic clusters from 1968 to 2022.

Discussion

Interestingly, our predicted antigenic map aligns closely with the antigenic map generated with serological data. Thus, our method is a promising tool for detecting antigenic variants and guiding the selection of vaccine candidates.

Emergence of Highly Pathogenic Avian Influenza A Virus (H5N1) of Clade 2.3.4.4b in Egypt, 2021-2022

Wild migratory birds have the capability to spread avian influenza virus (AIV) over long distances as well as transmit the virus to domestic birds. In this study, swab and tissue samples were obtained from 190 migratory birds during close surveillance in Egypt in response to the recent outbreaks of the highly pathogenic avian influenza (HPAI) H5N1 virus. The collected samples were tested for a variety of AIV subtypes (H5N1, H9N2, H5N8, and H6N2) as well as other pathogens such as NDV, IBV, ILT, IBDV, and WNV. Among all of the tested samples, the HPAI H5N1 virus was found in six samples; the other samples were found to be negative for all of the tested pathogens. The Egyptian HPAI H5N1 strains shared genetic traits with the HPAI H5N1 strains that are currently being reported in Europe, North America, Asia, and Africa in 2021-2022. Whole genome sequencing revealed markers associated with mammalian adaption and virulence traits among different gene segments, similar to those found in HPAI H5N1 strains detected in Europe and Africa. The detection of the HPAI H5N1 strain of clade 2.3.4.4b in wild birds in Egypt underlines the risk of the introduction of this strain into the local poultry population. Hence, there is reason to be vigilant and continue epidemiological and molecular monitoring of the AIV in close proximity to the domestic-wild bird interface.

Molecular Epidemiology and Evolutionary Analysis of Avian Influenza A(H5) Viruses Circulating in Egypt, 2019–2021

The highly pathogenic avian influenza (HPAI) H5N8 virus was first detected in Egypt in late 2016. Since then, the virus has spread rapidly among different poultry sectors, becoming the dominant HPAI H5 subtype reported in Egypt. Different genotypes of the HPAI H5N8 virus were reported in Egypt; however, the geographic patterns and molecular evolution of the Egyptian HPAI H5N8 viruses are still unclear. Here, extensive epidemiological surveillance was conducted, including more than half a million samples collected from different poultry sectors (farms/backyards/live bird markets) from all governorates in Egypt during 2019–2021. In addition, genetic characterization and evolutionary analyses were performed using 47 selected positive H5N8 isolates obtained during the same period. The result of the conducted surveillance showed that HPAI H5N8 viruses of clade 2.3.4.4b continue to circulate in different locations in Egypt, with an obvious seasonal pattern, and no further detection of the HPAI H5N1 virus of clade 2.2.1.2 was observed in the poultry population during 2019–2021. In addition, phylogenetic and Bayesian analyses revealed that two major genotypes (G5 and G6) of HPAI H5N8 viruses were continually expanding among the poultry sectors in Egypt. Notably, molecular dating analysis suggested that the Egyptian HPAI H5N8 virus is the potential ancestral viruses of the European H5N8 viruses of 2020–2021. In summary, the data of this study highlight the current epidemiology, diversity, and evolution of HPAI H5N8 viruses in Egypt and call for continuous monitoring of the genetic features of the avian influenza viruses in Egypt.

Antigenic characterization of influenza and SARS-CoV-2 viruses

Antigenic characterization of emerging and re-emerging viruses is necessary for the prevention of and response to outbreaks, evaluation of infection mechanisms, understanding of virus evolution, and selection of strains for vaccine development. Primary analytic methods, including enzyme-linked immunosorbent/lectin assays, hemagglutination inhibition, neuraminidase inhibition, micro-neutralization assays, and antigenic cartography, have been widely used in the field of influenza research. These techniques have been improved upon over time for increased analytical capacity, and some have been mobilized for the rapid characterization of the SARS-CoV-2 virus as well as its variants, facilitating the development of highly effective vaccines within 1 year of the initially reported outbreak. While great strides have been made for evaluating the antigenic properties of these viruses, multiple challenges prevent efficient vaccine strain selection and accurate assessment. For influenza, these barriers include the requirement for a large virus quantity to perform the assays, more than what can typically be provided by the clinical samples alone, cell- or egg-adapted mutations that can cause antigenic mismatch between the vaccine strain and circulating viruses, and up to a 6-month duration of vaccine development after vaccine strain selection, which allows viruses to continue evolving with potential for antigenic drift and, thus, antigenic mismatch between the vaccine strain and the emerging epidemic strain. SARS-CoV-2 characterization has faced similar challenges with the additional barrier of the need for facilities with high biosafety levels due to its infectious nature. In this study, we review the primary analytic methods used for antigenic characterization of influenza and SARS-CoV-2 and discuss the barriers of these methods and current developments for addressing these challenges.

Next-Generation Computationally Designed Influenza Hemagglutinin Vaccines Protect against H5Nx Virus Infections

H5N1 COBRA hemagglutinin (HA) sequences, termed human COBRA-2 HA, were constructed through layering of HA sequences from viruses isolated from humans collected between 2004-2007 using only clade 2 strains. These COBRA HA proteins, when expressed on the surface of virus-like particles (VLP), elicited protective immune responses in mice, ferrets, and non-human primates. However, these vaccines were not as effective at inducing neutralizing antibodies against newly circulating viruses. Therefore, COBRA HA-based vaccines were updated in order to elicit protective antibodies against the current circulating clades of H5Nx viruses. Next-generation COBRA HA vaccines were designed to encompass the newly emerging viruses circulating in wild avian populations. HA amino acid sequences from avian and human H5 influenza viruses isolated between 2011-2017 were downloaded from the GISAID (Global Initiative on Sharing All Influenza Data). Mice were vaccinated with H5 COBRA rHA that elicited antibodies with hemagglutinin inhibition (HAI) activity against H5Nx viruses from five clades. The H5 COBRA rHA vaccine, termed IAN8, elicited protective immune responses against mice challenged with A/Sichuan/26621/2014 and A/Vietnam/1203/2004. This vaccine elicited antibodies with HAI activity against viruses from clades 2.2, 2.3.2.1, 2.3.4.2, 2.2.1 and 2.2.2. Lungs from vaccinated mice had decreased viral titers and the levels of cellular infiltration in mice vaccinated with IAN-8 rHA were similar to mice vaccinated with wild-type HA comparator vaccines or mock vaccinated controls. Overall, these next-generation H5 COBRA HA vaccines elicited protective antibodies against both historical H5Nx influenza viruses, as well as currently circulating clades of H5N1, H5N6, and H5N8 influenza viruses.

Triple reassortment increases compatibility among viral ribonucleoprotein genes of contemporary avian and human influenza A viruses

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.

Variations outside the conserved motifs of PB1 catalytic active site may affect replication efficiency of the RNP complex of influenza A virus

PB1 functions as the catalytic subunit of influenza virus RNA polymerase complex and plays an essential role in viral RNA transcription and replication. To determine plasticity in the PB1 enzymatic site and map catalytically important residues, 658 mutants were constructed, each with one to seven mutations in the enzymatic site of PB1. The polymerase activities of these mutants were quantified using a minigenome assay, and polymerase activity-associated residues were identified using sparse learning. Results showed that polymerase activities are affected by the residues not only within the conserved motifs, but also across the inter-motif regions of PB1, and the latter are primarily located at the base of the palm domain, a region that is conserved in avian PB1 but with high sequence diversity in swine PB1. Our results suggest that mutations outside the PB1 conserved motifs may affect RNA replication and could be associated with influenza virus host adaptation.

Pathological and Molecular Characterization of H5 Avian Influenza Virus in Poultry Flocks from Egypt over a Ten-Year Period (2009-2019)

Simple Summary

Avian influenza virus (H5) remains one of the challenging zoonotic viruses in Egypt. Our study investigated the occurrence of this virus among chickens from Dakhalia governorate, Egypt over ten years through histopathological examination and molecular characterization of the virus. The molecular characterization was followed by sequencing and phylogenetic analysis of the positive samples. Importantly, we have reported several interesting pathological changes and high occurrence of the H5 avian influenza virus, the phylogenetic analysis revealed that positive samples were aligned with several Egyptian sub clades. Clearly, our study concludes the widespread of the virus among poultry flocks in Egypt and suggests further future research aims to develop an efficient surveillance program with investigation into the effectiveness of the implemented control measures for controlling this disease of public health concern.

Abstract

Avian influenza virus (AIV) remains one of the enzootic zoonotic diseases that challenges the poultry industry in Egypt. In the present study, a total of 500 tissue samples were collected from 100 chicken farms (broilers and layers) suspected to be infected with AIV through the period from 2009 to 2019 from Dakahlia governorate, Egypt. These samples were pooled in 100 working samples and screened for AIV then the positive samples were subjected to histopathological examination combined with real time-polymerase chain reaction (RRT-PCR). RRT-PCR positive samples were also subjected to conventional reverse transcriptase-polymerase chain reaction (RT-PCR) for detection of H5 AIV and some of these resulting positive samples were sequenced for detection of the molecular nature of the studied virus. Interestingly, the histopathological examination revealed necrotic liver with leukocytic infiltration with degenerative changes with necrotic pancreatitis, edema, and intense lymphoid depletion of splenic tissue and hyperplastic tracheal epithelium. Likewise, edema and congested sub mucosal blood vessels and intense bronchial necrosis with hyalinized wall vascular wall and heterophils infiltration were reported. Pneumonic areas with intense leukocytic aggregation mainly and vasculitis of the pulmonary blood vessels were also detected in lung. Collectively, these significant pathological changes in examined tissues cohered with AIV infection. Regarding the molecular characterization, 66 samples were positive for AIV by RRT-PCR and 52 of them were positive for H5 AIV by RT-PCR. The phylogenetic analysis revealed that the H5 viruses identified in this study were aligned with other Egyptian H5N1 AIVs in the Egyptian sub clade 2.2.1, while some of the identified strains were aligned with other Egyptian H5N8 strains in the new Egyptian sub clade 2.3.4.4. Taken together, our present findings emphasize the wide spread of AIV in Egypt and the importance of developing an efficient surveillance and periodical screening program for controlling such disease of public health concern.

Multi-task learning sparse group lasso: a method for quantifying antigenicity of influenza A(H1N1) virus using mutations and variations in glycosylation of Hemagglutinin

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.

Human COBRA 2 vaccine contains two major epitopes that are responsible for eliciting neutralizing antibody responses against heterologous clades of viruses

Highly pathogenic H5N1 influenza viruses continue to spread around the globe and reassort with low pathogenic avian influenza viruses often resulting in morbidity and mortality to not only waterfowl, but also poultry. Our group previously developed two hemagglutinin (HA) based vaccines using a methodology termed computationally optimized broadly reactive antigen (COBRA). Each of these HA antigens, Human COBRA 2 (Hu-CO) and Human-Avian COBRA 2 (Hu-Av CO) elicit antibodies with hemagglutination-inhibition (HAI) activity against viruses from various clades, but not always the same viruses. Here, we have sought to identify residues in these two HA molecules that are critical fordifferential HAI activity against various H5Nx influenza viruses. The two HA antigens are remarkedly similar in the globular head region, except for 4 residues at amino acids 140, 141, 155, and 156. By mutating these amino acids in each HA antigen, chimeric HA proteins were used to elicit immune responses in mice. When the Asn-Thr pair at position 155-156 in the Hu-CO HA was converted to the Ser-Ala residues found in the Hu-Av CO HA, the elicited antibodies lost HAI activity against clade 2.3.2.1 H5Nx viruses, such as A/Hubei/01/2010. When this Asn-Thr motif was added at these positions in the Hu-Av CO HA molecule, HAI activity in the elicited sera against A/Hubei/01/2010 was significantly increased. We speculate that a putative N-linked glycosylation at this location in the Hu-CO HA antigen is a key driver in the elicitation of antibodies with HAI activity to different locations on wild-type H5 HA molecules resulting in differential neutralization of viral infection and protection in vivo against H5 influenza virus induced disease.

Graph-guided multi-task sparse learning model: a method for identifying antigenic variants of influenza A(H3N2) virus

Motivation

Influenza virus antigenic variants continue to emerge and cause disease outbreaks. Time-consuming, costly and middle-throughput serologic methods using virus isolates are routinely used to identify influenza antigenic variants for vaccine strain selection. However, the resulting data are notoriously noisy and difficult to interpret and integrate because of variations in reagents, supplies and protocol implementation. A novel method without such limitations is needed for antigenic variant identification.

Results

We developed a Graph-Guided Multi-Task Sparse Learning (GG-MTSL) model that uses multi-sourced serologic data to learn antigenicity-associated mutations and infer antigenic variants. By applying GG-MTSL to influenza H3N2 hemagglutinin sequences, we showed the method enables rapid characterization of antigenic profiles and identification of antigenic variants in real time and on a large scale. Furthermore, sequences can be generated directly by using clinical samples, thus minimizing biases due to culture-adapted mutation during virus isolation.

Availability and implementation

MATLAB source codes developed for GG-MTSL are available through http://sysbio.cvm.msstate.edu/files/GG-MTSL/.

Supplementary information

Supplementary data are available at Bioinformatics online.

Detection of Microaneurysms in Fundus Images Based on an Attention Mechanism

Microaneurysms (MAs) are the earliest detectable diabetic retinopathy (DR) lesions. Thus, the ability to automatically detect MAs is critical for the early diagnosis of DR. However, achieving the accurate and reliable detection of MAs remains a significant challenge due to the size and complexity of retinal fundus images. Therefore, this paper presents a novel MA detection method based on a deep neural network with a multilayer attention mechanism for retinal fundus images. First, a series of equalization operations are performed to improve the quality of the fundus images. Then, based on the attention mechanism, multiple feature layers with obvious target features are fused to achieve preliminary MA detection. Finally, the spatial relationships between MAs and blood vessels are utilized to perform a secondary screening of the preliminary test results to obtain the final MA detection results. We evaluated the method on the IDRiD_VOC dataset, which was collected from the open IDRiD dataset. The results show that our method effectively improves the average accuracy and sensitivity of MA detection.

Predicting Antigenicity of Influenza A Viruses Using biophysical ideas

Antigenic variations of influenza A viruses are induced by genomic mutation in their trans-membrane protein HA1, eliciting viral escape from neutralization by antibodies generated in prior infections or vaccinations. Prediction of antigenic relationships among influenza viruses is useful for designing (or updating the existing) influenza vaccines, provides important insights into the evolutionary mechanisms underpinning viral antigenic variations, and helps to understand viral epidemiology. In this study, we present a simple and physically interpretable model that can predict antigenic relationships among influenza A viruses, based on biophysical ideas, using both genomic amino acid sequences and experimental antigenic data. We demonstrate the applicability of the model using a benchmark dataset of four subtypes of influenza A (H1N1, H3N2, H5N1, and H9N2) viruses and report on its performance profiles. Additionally, analysis of the model’s parameters confirms several observations that are consistent with the findings of other previous studies, for which we provide plausible explanations.

Respiratory disease due to mixed viral infections in poultry flocks in Egypt between 2017 and 2018: Upsurge of highly pathogenic avian influenza virus subtype H5N8 since 2018

For several years, poultry production in Egypt has been suffering from co-circulation of multiple respiratory viruses including highly pathogenic avian influenza virus (HPAIV) H5N1 (clade 2.2.1.2) and low pathogenic H9N2 (clade G1-B). Incursion of HPAIV H5N8 (clade 2.3.4.4b) to Egypt in November 2016 via wild birds followed by spread into commercial poultry flocks further complicated the situation. Current analyses focussed on 39 poultry farms suffering from respiratory manifestation and high mortality in six Egyptian governorates during 2017-2018. Real-time RT-PCR (RT-qPCR) substantiated the co-presence of at least two respiratory virus species in more than 80% of the investigated flocks. The percentage of HPAIV H5N1-positive holdings was fairly stable in 2017 (12.8%) and 2018 (10.2%), while the percentage of HPAIV H5N8-positive holdings increased from 23% in 2017 to 66.6% during 2018. The proportion of H9N2-positive samples was constantly high (2017:100% and 2018:63%), and H9N2 co-circulated with HPAIV H5N8 in 22 out of 39 (56.8%) flocks. Analyses of 26 H5, 18 H9 and 4 N2 new sequences confirmed continuous genetic diversification. In silico analysis revealed numerous amino acid substitutions in the HA and NA proteins suggestive of increased adaptation to mammalian hosts and putative antigenic variation. For sensitive detection of H9N2 viruses by RT-qPCR, an update of primers and probe sequences was crucial. Reasons for the relative increase of HPAIV H5N8 infections versus H5N1 remained unclear, but lack of suitable vaccines against clade 2.3.4.4b cannot be excluded. A reconsideration of surveillance and control measures should include updating of diagnostic tools and vaccination strategies.

Multiple introductions of reassorted highly pathogenic avian influenza viruses (H5N8) clade 2.3.4.4b causing outbreaks in wild birds and poultry in Egypt

Recently, an increased incidence of outbreaks of highly pathogenic avian influenza (HPAI) H5N8 in poultry linked to infected migratory birds has been reported from different European, Asian and African countries. In Egypt, incursion of HPAI H5N8 virus of clade 2.3.4.4b has been recently registered. Full genomic characterization of 3 virus isolates from wild birds and poultry (backyard and commercial farm sectors) showed high nucleotide similarity among the HA, NA, M, and NS gene segments of the three Egyptian HPAI H5N8 viruses, indicating that they are descendants of a common ancestral virus. However, the analyzed Egyptian H5N8 viruses revealed distinct genotypes involving different origins of the PB2, PB1, PA and/or NP segments. In genotype-1 represented by strain A/common-coot/Egypt/CA285/2016 the PB2 and NP segments showed closest relationship to H5N6 and H6N2 viruses, recently detected in Italy. The second is replacement of PB1 and NP genes A novel reassortant, represented by strain A/duck/Egypt/SS19/2017, showed an exchange of PB1 and NP genes which might have originated from H6N8 or H1N1 and H6N2 viruses. Finally, replacement of PA and NP genes characterized strain A/duck/Egypt/F446/2017. Bayesian phylogeographic analyses revealed that Egyptian H5N8 viruses are highly likely derived from Russian 2016 HPAI H5N8 virus (A/great_crested_grebe/Uvs-Nuur_Lake/341/2016 (H5N8)) and the reassortment likely occurred before incursion to Egypt.

A Sparse Learning Framework for Joint Effect Analysis of Copy Number Variants

Copy number variants (CNVs), including large deletions and duplications, represent an unbalanced change of DNA segments. Abundant in human genomes, CNVs contribute to a large proportion of human genetic diversity, with impact on many human phenotypes. Although recent advances in genetic studies have shed light on the impact of individual CNVs on different traits, the analysis of joint effect of multiple interactive CNVs lags behind from many perspectives. A primary reason is that the large number of CNV combinations and interactions in the human genome make it computationally challenging to perform such joint analysis. To address this challenge, we developed a novel sparse learning framework that combines sparse learning with biological networks to identify interacting CNVs with joint effect on particular traits. We showed that our approach performs well in identifying CNVs with joint phenotypic effect using simulated data. Applied to a real human genomic dataset from the 1,000 Genomes Project, our approach identified multiple CNVs that collectively contribute to population differentiation. We found a set of multiple CNVs that have joint effect in different populations, and affect gene expression differently in distinct populations. These results provided a collection of CNVs that likely have downstream biomedical implications in individuals from diverse population backgrounds.

Predicting influenza antigenicity from Hemagglutintin sequence data based on a joint random forest method

Timely identification of emerging antigenic variants is critical to influenza vaccine design. The accuracy of a sequence-based antigenic prediction method relies on the choice of amino acids substitution matrices. In this study, we first compared a comprehensive 95 substitution matrices reflecting various amino acids properties in predicting the antigenicity of influenza viruses by a random forest model. We then proposed a novel algorithm called joint random forest regression (JRFR) to jointly consider top substitution matrices. We applied JRFR to human H3N2 seasonal influenza data from 1968 to 2003. A 10-fold cross-validation shows that JRFR outperforms other popular methods in predicting antigenic variants. In addition, our results suggest that structure features are most relevant to influenza antigenicity. By restricting the analysis to data involving two adjacent antigenic clusters, we inferred a few key amino acids mutation driving the 11 historical antigenic drift events, pointing to experimentally validated mutations. Finally, we constructed an antigenic cartography of all H3N2 viruses with hemagglutinin (the glycoprotein on the surface of the influenza virus responsible for its binding to host cells) sequence available from NCBI flu database, and showed an overall correspondence and local inconsistency between genetic and antigenic evolution of H3N2 influenza viruses.

Searching Genome-Wide Multi-Locus Associations for Multiple Diseases Based on Bayesian Inference

Taking the advantage of high-throughput single nucleotide polymorphism (SNP) genotyping technology, large genome-wide association studies (GWASs) have been considered to hold promise for unraveling complex relationships between genotypes and phenotypes. Current multi-locus-based methods are insufficient to detect interactions with diverse genetic effects on multifarious diseases. Also, statistic tests for high-order epistasis ( ≥ 2 SNPs) raise huge computational and analytical challenges because the computation increases exponentially as the growth of the cardinality of SNPs combinations. In this paper, we provide a simple, fast and powerful method, named DAM, using Bayesian inference to detect genome-wide multi-locus epistatic interactions in multiple diseases. Experimental results on simulated data demonstrate that our method is powerful and efficient. We also apply DAM on two GWAS datasets from WTCCC, i.e., Rheumatoid Arthritis and Type 1 Diabetes, and identify some novel findings. Therefore, we believe that our method is suitable and efficient for the full-scale analysis of multi-disease-related interactions in GWASs.

Different cross protection scopes of two avian influenza H5N1 vaccines against infection of layer chickens with a heterologous highly pathogenic virus

Avian influenza (AI) virus strains vary in antigenicity, and antigenic differences between circulating field virus and vaccine virus will affect the effectiveness of vaccination of poultry. Antigenic relatedness can be assessed by measuring serological cross-reactivity using haemagglutination inhibition (HI) tests. Our study aims to determine the relation between antigenic relatedness expressed by the Archetti-Horsfall ratio, and reduction of virus transmission of highly pathogenic H5N1 AI strains among vaccinated layers. Two vaccines were examined, derived from H5N1 AI virus strains A/Ck/WJava/Sukabumi/006/2008 and A/Ck/CJava/Karanganyar/051/2009. Transmission experiments were carried out in four vaccine and two control groups, with six sets of 16 specified pathogen free (SPF) layer chickens. Birds were vaccinated at 4weeks of age with one strain and challenge-infected with the homologous or heterologous strain at 8weeks of age. No transmission or virus shedding occurred in groups challenged with the homologous strain. In the group vaccinated with the Karanganyar strain, high cross-HI responses were observed, and no transmission of the Sukabumi strain occurred. However, in the group vaccinated with the Sukabumi strain, cross-HI titres were low, virus shedding was not reduced, and multiple transmissions to contact birds were observed. This study showed large differences in cross-protection of two vaccines based on two different highly pathogenic H5N1 virus strains. This implies that extrapolation of in vitro data to clinical protection and reduction of virus transmission might not be straightforward.

Reassortant highly pathogenic influenza A(H5N6) virus in Laos

In March 2014, avian influenza in poultry in Laos was caused by an emergent influenza A(H5N6) virus. Genetic analysis indicated that the virus had originated from reassortment of influenza A(H5N1) clade 2.3.2.1b, variant clade 2.3.4, and influenza A(H6N6) viruses that circulate broadly in duck populations in southern and eastern China.

Broad cross-reactive epitopes of the H5N1 influenza virus identified by murine antibodies against the A/Vietnam/1194/2004 hemagglutinin

There is an urgent need for a rapid diagnostic system to detect the H5 subtype of the influenza A virus. We previously developed monoclonal antibodies (mAbs) against the H5 hemagglutinin (HA) for use in a rapid diagnostic kit. In this study, we determined the epitopes of the anti-H5 HA murine mAbs OM-b, AY-2C2, and YH-1A1. Binding assays of the mAbs to different strains of H5 HAs indicated that OM-b and AY-2C2 cross-reacted with HAs from clades 1, 2.1.3.2, 2.2, and 2.3.4, whereas YH-1A1 failed to bind to those of clades 2.1.3.2 and 2.3.4. HA chimeras revealed that the epitopes for each of the mAbs were in the HA1 region. Analysis of escape mutants revealed that OM-b and AY-2C2 mAbs interacted mainly with amino acid residues D43 and G46, and the YH-1A1 mAb interacted with G139 and K or R140 of H5 HA. Multiple alignments of H5 HA protein sequences showed that D43 and G46 were very conserved among H5N1 HAs, except those in clade 2.2.1 and clade 7 (88.7%). The epitope for YH-1A1 mAb was highly variable in the HAs of H5N1, although it was well conserved in those of H5N2-N9. The OM-b and AY-2C2 mAbs could bind to the HAs of clades 1.1 and 2.3.2.1 that are currently epidemic in Asia, and we conclude that these would be effective for the detection of H5N1 infections in this region.

Molecular characterization of avian influenza H5N1 virus in Egypt and the emergence of a novel endemic subclade

Clade 2.2 highly pathogenic H5N1 viruses have been in continuous circulation in Egyptian poultry since 2006. Their persistence caused significant genetic drift that led to the reclassification of these viruses into subclades 2.2.1 and 2.2.1.1. Here, we conducted full-genome sequence and phylogenetic analyses of 45 H5N1 isolated during 2006-2013 through systematic surveillance in Egypt, and 53 viruses that were sequenced previously and available in the public domain. Results indicated that H5N1 viruses in Egypt continue to evolve and a new distinct cluster has emerged. Mutations affecting viral virulence, pathogenicity, transmission, receptor-binding preference and drug resistance were studied. In light of our findings that H5N1 in Egypt continues to evolve, surveillance and molecular studies need to be sustained.

Mutation from arginine to lysine at the position 189 of hemagglutinin contributes to the antigenic drift in H3N2 swine influenza viruses

Two distinct antigenic clusters were previously identified among the H3N2 swine influenza A viruses (IAVs) and were designated H3N2SIV-alpha and H3N2SIV-beta (Feng et al., 2013. Journal of Virology 87 (13), 7655-7667). A consistent mutation was observed at the position 189 of hemagglutinin (R189K) between H3N2SIV-alpha and H3N2SIV-beta fair isolates. To evaluate the contribution of R189K mutation to the antigenic drift from H3N2SIV-alpha to H3N2SIV-beta, four reassortant viruses with 189R or 189K were generated. The antigenic cartography demonstrated that the R189K mutation in the hemagglutinin of H3N2 IAV contributed to the antigenic drift, separating these viruses into H3N2SIV-alpha to H3N2SIV-beta. This R189K mutation was also found to contribute to the cross-reaction with several ferret sera raised against historical human IAVs with hemagglutinin carrying 189K. This study suggests that the R189K mutation plays a vital role in the antigenicity of swine and human H3N2 IAVs and identification of this antigenic determinant will help us rapidly identify antigenic variants in influenza surveillance.

Structural and antigenic variation among diverse clade 2 H5N1 viruses

Antigenic variation among circulating H5N1 highly pathogenic avian influenza A viruses mandates the continuous production of strain-specific pre-pandemic vaccine candidates and represents a significant challenge for pandemic preparedness. Here we assessed the structural, antigenic and receptor-binding properties of three H5N1 HPAI virus hemagglutinins, which were recently selected by the WHO as vaccine candidates [A/Egypt/N03072/2010 (Egypt10, clade 2.2.1), A/Hubei/1/2010 (Hubei10, clade 2.3.2.1) and A/Anhui/1/2005 (Anhui05, clade 2.3.4)]. These analyses revealed that antigenic diversity among these three isolates was restricted to changes in the size and charge of amino acid side chains at a handful of positions, spatially equivalent to the antigenic sites identified in H1 subtype viruses circulating among humans. All three of the H5N1 viruses analyzed in this study were responsible for fatal human infections, with the most recently-isolated strains, Hubei10 and Egypt10, containing multiple residues in the receptor-binding site of the HA, which were suspected to enhance mammalian transmission. However, glycan-binding analyses demonstrated a lack of binding to human α2-6-linked sialic acid receptor analogs for all three HAs, reinforcing the notion that receptor-binding specificity contributes only partially to transmissibility and pathogenesis of HPAI viruses and suggesting that changes in host specificity must be interpreted in the context of the host and environmental factors, as well as the virus as a whole. Together, our data reveal structural linkages with phylogenetic and antigenic analyses of recently emerged H5N1 virus clades and should assist in interpreting the significance of future changes in antigenic and receptor-binding properties.

Antigenic analysis of H5N1 highly pathogenic avian influenza viruses circulating in Egypt (2006-2012)

The highly pathogenic avian influenza (HPAI) H5N1 in Egypt circulated continuously after its introduction in February 2006 with substantial economic losses and frequent human infections. Phylogenetic analysis of the available HA sequences revealed the presence of two main sublineages; the classic 2.2.1 and the variant 2.2.1.1. The classic 2.2.1 had subdivided into two clusters of viruses; cluster C1 contained the originally introduced virus and isolates from 2006 to 2009 and cluster C2 emerged in 2007 and continues to circulate. The variant 2.2.1.1 represents the isolates mainly from chickens and subdivided into two clusters; cluster V1 contains isolates from 2007 to 2009 and cluster V2 contains isolates from 2008 to 2011. Sequence analysis revealed 28 amino acid mutations in the previously reported antigenic sites and high evolution rate which may be due to selective pressure from vaccination and/or natural infection. Antigenic analysis of 18 H5N1 isolates from 2006 to 2012 that represent different clusters was conducted using hemagglutination inhibition (HI) and virus neutralization (VN) assays using hyperimmune sera produced by immunizing SPF chickens with inactivated whole-virus. Antigenic relatedness of ancestral Egyptian H5N1 isolate (459-3/06) with other isolates ranged from 30.7% to 79.1% indicating significant antigenic drift of the H5N1 viruses from the ancestral strains. The antigenic relatedness between C2 and V2 clusters ranged from 28.9% to 68% supporting the need for vaccine seed strains from both clusters. Interestingly, A/CK/EG/1709-6/2008 H5N1 strain showed a broad cross reactivity against viruses in different H5N1 clusters (antigenic relatedness ranged from 63.9% to 85.8%) demonstrating a potential candidate as a vaccine strain. Antigenic cartography which facilitates a quantitative interpretation and easy visualization of serological data was constructed based on HI results and further demonstrated the several antigenic groups among Egyptian H5N1 viruses. In conclusion, the cross reactivity between the co-circulating H5N1 strains may not be adequate for protection against each other and it is recommended to test vaccines that contain isolates from different antigenic groups in experimental infection trials for the selection of vaccine seed strain. Furthermore, the continuous monitoring for detecting the emerging variants followed by detailed antigenic analysis for updating vaccines is warranted.

Diabetic retinopathy risk prediction for fundus examination using sparse learning: a cross-sectional study

BACKGROUND

Blindness due to diabetic retinopathy (DR) is the major disability in diabetic patients. Although early management has shown to prevent vision loss, diabetic patients have a low rate of routine ophthalmologic examination. Hence, we developed and validated sparse learning models with the aim of identifying the risk of DR in diabetic patients.

METHODS

Health records from the Korea National Health and Nutrition Examination Surveys (KNHANES) V-1 were used. The prediction models for DR were constructed using data from 327 diabetic patients, and were validated internally on 163 patients in the KNHANES V-1. External validation was performed using 562 diabetic patients in the KNHANES V-2. The learning models, including ridge, elastic net, and LASSO, were compared to the traditional indicators of DR.

RESULTS

Considering the Bayesian information criterion, LASSO predicted DR most efficiently. In the internal and external validation, LASSO was significantly superior to the traditional indicators by calculating the area under the curve (AUC) of the receiver operating characteristic. LASSO showed an AUC of 0.81 and an accuracy of 73.6% in the internal validation, and an AUC of 0.82 and an accuracy of 75.2% in the external validation.

CONCLUSION

The sparse learning model using LASSO was effective in analyzing the epidemiological underlying patterns of DR. This is the first study to develop a machine learning model to predict DR risk using health records. LASSO can be an excellent choice when both discriminative power and variable selection are important in the analysis of high-dimensional electronic health records.

Development of a magnetic electrochemical bar code array for point mutation detection in the H5N1 neuraminidase gene

Since its first official detection in the Guangdong province of China in 1996, the highly pathogenic avian influenza virus of H5N1 subtype (HPAI H5N1) has reportedly been the cause of outbreaks in birds in more than 60 countries, 24 of which were European. The main issue is still to develop effective antiviral drugs. In this case, single point mutation in the neuraminidase gene, which causes resistance to antiviral drug and is, therefore, subjected to many studies including ours, was observed. In this study, we developed magnetic electrochemical bar code array for detection of single point mutations (mismatches in up to four nucleotides) in H5N1 neuraminidase gene. Paramagnetic particles Dynabeads® with covalently bound oligo (dT)₂₅ were used as a tool for isolation of complementary H5N1 chains (H5N1 Zhejin, China and Aichi). For detection of H5N1 chains, oligonucleotide chains of lengths of 12 (+5 adenine) or 28 (+5 adenine) bp labeled with quantum dots (CdS, ZnS and/or PbS) were used. Individual probes hybridized to target molecules specifically with efficiency higher than 60%. The obtained signals identified mutations present in the sequence. Suggested experimental procedure allows obtaining further information from the redox signals of nucleic acids. Moreover, the used biosensor exhibits sequence specificity and low limits of detection of subnanogram quantities of target nucleic acids.

Using sequence data to infer the antigenicity of influenza virus

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.

Vaccination with recombinant RNA replicon particles protects chickens from H5N1 highly pathogenic avian influenza virus

Highly pathogenic avian influenza viruses (HPAIV) of subtype H5N1 not only cause a devastating disease in domestic chickens and turkeys but also pose a continuous threat to public health. In some countries, H5N1 viruses continue to circulate and evolve into new clades and subclades. The rapid evolution of these viruses represents a problem for virus diagnosis and control. In this work, recombinant vesicular stomatitis virus (VSV) vectors expressing HA of subtype H5 were generated. To comply with biosafety issues the G gene was deleted from the VSV genome. The resulting vaccine vector VSV*ΔG(HA) was propagated on helper cells providing the VSV G protein in trans. Vaccination of chickens with a single intramuscular dose of 2×10⁸ infectious replicon particles without adjuvant conferred complete protection from lethal H5N1 infection. Subsequent application of the same vaccine strongly boosted the humoral immune response and completely prevented shedding of challenge virus and transmission to sentinel birds. The vaccine allowed serological differentiation of infected from vaccinated animals (DIVA) by employing a commercially available ELISA. Immunized chickens produced antibodies with neutralizing activity against multiple H5 viruses representing clades 1, 2.2, 2.5, and low-pathogenic avian influenza viruses (classical clade). Studies using chimeric H1/H5 hemagglutinins showed that the neutralizing activity was predominantly directed against the globular head domain. In summary, these results suggest that VSV replicon particles are safe and potent DIVA vaccines that may help to control avian influenza viruses in domestic poultry.

Biological properties of H5 hemagglutinin expressed by vaccinia virus vector and its immunological reactivity with human sera

A recombinant vaccinia virus harboring the full length hemagglutinin (HA) gene derived from a highly pathogenic avian influenza A/Thailand/1(KAN-1)/2004 (H5N1) virus (rVac-H5 HA virus) was constructed. The immunogenicity of the expressed HA protein was characterized using goat antiserum, mouse monoclonal antibody, and human sera. The expressed HA protein localized both in the cytoplasm and on the cytoplasmic membrane of the thymidine kinase negative cells infected with the rVac-H5 HA virus, as determined by immunofluorescence assay. Western blot analysis demonstrated that the rVac-H5 HA protein was post-translationally processed by proteolytic cleavage of the HA0 precursor into HA1 and HA2 domains; and all of these HA forms were immunogenic in BALB/c mice. The molecular weight (MW) of each HA domain was the same as the wild-type H5 HA produced in Madin-Darby canine kidney cells infected with the H5N1 virus, but was higher than that expressed by a baculovirus-insect cell system. Sera from all H5N1 survivors reacted to HA0, HA1, and HA2 domains; whereas sera from H5N1-uninfected subjects reacted to the HA2 domain only, but not to HA0 or HA1, indicating that some cross-subtypic immunity exists in the general population. There was a lot-to-lot variation of the recombinant HA produced in the baculovirus-insect cell system that might affect the detection rate of antibody directed against certain HA domains.