Putative amino acid determinants of the emergence of the 2009 influenza A (H1N1) virus in the human population

Influenza virus genotype to phenotype predictions through machine learning: a systematic review

Background: There is great interest in understanding the viral genomic predictors of phenotypic traits that allow influenza A viruses to adapt to or become more virulent in different hosts. Machine learning techniques have demonstrated promise in addressing this critical need for other pathogens because the underlying algorithms are especially well equipped to uncover complex patterns in large datasets and produce generalizable predictions for new data. As the body of research where these techniques are applied for influenza A virus phenotype prediction continues to grow, it is useful to consider the strengths and weaknesses of these approaches to understand what has prevented these models from seeing widespread use by surveillance laboratories and to identify gaps that are underexplored with this technology. Methods and Results: We present a systematic review of English literature published through 15 April 2021 of studies employing machine learning methods to generate predictions of influenza A virus phenotypes from genomic or proteomic input. Forty-nine studies were included in this review, spanning the topics of host discrimination, human adaptability, subtype and clade assignment, pandemic lineage assignment, characteristics of infection, and antiviral drug resistance. Conclusions: Our findings suggest that biases in model design and a dearth of wet laboratory follow-up may explain why these models often go underused. We, therefore, offer guidance to overcome these limitations, aid in improving predictive models of previously studied influenza A virus phenotypes, and extend those models to unexplored phenotypes in the ultimate pursuit of tools to enable the characterization of virus isolates across surveillance laboratories.

Changes to the dynamic nature of hemagglutinin and the emergence of the 2009 pandemic H1N1 influenza virus

The virologic factors that limit the transmission of swine influenza viruses between humans are unresolved. While it has been shown that acquisition of the neuraminidase (NA) and matrix (M) gene segments from a Eurasian-lineage swine virus was required for airborne transmission of the 2009 pandemic H1N1 virus (H1N1pdm09), we show here that an arginine to lysine change in the hemagglutinin (HA) was also necessary. This change at position 149 was distal to the receptor binding site but affected virus-receptor affinity and HA dynamics, allowing the virus to replicate more efficiently in nasal turbinate epithelium and subsequently transmit between ferrets. Receptor affinity should be considered as a factor limiting swine virus spread in humans.

Knowledge discovery and sequence-based prediction of pandemic influenza using an integrated classification and association rule mining (CBA) algorithm

Pandemic influenza is a major concern worldwide. Availability of advanced technologies and the nucleotide sequences of a large number of pandemic and non-pandemic influenza viruses in 2009 provide a great opportunity to investigate the underlying rules of pandemic induction through data mining tools. Here, for the first time, an integrated classification and association rule mining algorithm (CBA) was used to discover the rules underpinning alteration of non-pandemic sequences to pandemic ones. We hypothesized that the extracted rules can lead to the development of an efficient expert system for prediction of influenza pandemics. To this end, we used a large dataset containing 5373 HA (hemagglutinin) segments of the 2009 H1N1 pandemic and non-pandemic influenza sequences. The analysis was carried out for both nucleotide and protein sequences. We found a number of new rules which potentially present the undiscovered antigenic sites at influenza structure. At the nucleotide level, alteration of thymine (T) at position 260 was the key discriminating feature in distinguishing non-pandemic from pandemic sequences. At the protein level, rules including I233K, M334L were the differentiating features. CBA efficiently classifies pandemic and non-pandemic sequences with high accuracy at both the nucleotide and protein level. Finding hotspots in influenza sequences is a significant finding as they represent the regions with low antibody reactivity. We argue that the virus breaks host immunity response by mutation at these spots. Based on the discovered rules, we developed the software, "Prediction of Pandemic Influenza" for discrimination of pandemic from non-pandemic sequences. This study opens a new vista in discovery of association rules between mutation points during evolution of pandemic influenza.

K-Pax2: Bayesian identification of cluster-defining amino acid positions in large sequence datasets

The recent growth in publicly available sequence data has introduced new opportunities for studying microbial evolution and spread. Because the pace of sequence accumulation tends to exceed the pace of experimental studies of protein function and the roles of individual amino acids, statistical tools to identify meaningful patterns in protein diversity are essential. Large sequence alignments from fast-evolving micro-organisms are particularly challenging to dissect using standard tools from phylogenetics and multivariate statistics because biologically relevant functional signals are easily masked by neutral variation and noise. To meet this need, a novel computational method is introduced that is easily executed in parallel using a cluster environment and can handle thousands of sequences with minimal subjective input from the user. The usefulness of this kind of machine learning is demonstrated by applying it to nearly 5000 haemagglutinin sequences of influenza A/H3N2.Antigenic and 3D structural mapping of the results show that the method can recover the major jumps in antigenic phenotype that occurred between 1968 and 2013 and identify specific amino acids associated with these changes. The method is expected to provide a useful tool to uncover patterns of protein evolution.

Error-prone pcr-based mutagenesis strategy for rapidly generating high-yield influenza vaccine candidates

Vaccination is the primary strategy for the prevention and control of influenza outbreaks. However, the manufacture of influenza vaccine requires a high-yield seed strain, and the conventional methods for generating such strains are time consuming. In this study, we developed a novel method to rapidly generate high-yield candidate vaccine strains by integrating error-prone PCR, site-directed mutagenesis strategies, and reverse genetics. We used this method to generate seed strains for the influenza A(H1N1)pdm09 virus and produced six high-yield candidate strains. We used a mouse model to assess the efficacy of two of the six candidate strains as a vaccine seed virus: both strains provided complete protection in mice against lethal challenge, thus validating our method. Results confirmed that the efficacy of these candidate vaccine seed strains was not affected by the yield-optimization procedure.

Polymorphisms in the haemagglutinin gene influenced the viral shedding of pandemic 2009 influenza virus in swine

Interactions between the viral surface glycoprotein haemagglutinin (HA) and the corresponding receptors on host cells is one important aspect of influenza virus infection. Mutations in HA have been described to affect pathogenicity, antigenicity and the transmission of influenza viruses. Here, we detected polymorphisms present in HA genes of two pandemic 2009 H1N1 (H1N1pdm09) isolates, A/California/04/2009 (Ca/09) and A/Mexico/4108/2009 (Mx/09), that resulted in amino acid changes at positions 186 (S to P) and 194 (L to I) of the mature HA1 protein. Although not reported in the published H1N1pdm09 consensus sequence, the P186 genotype was more readily detected in primary infected and contact-naïve pigs when inoculated with a heterogeneous mixed stock of Ca/09. Using reverse genetics, we engineered Ca/09 and Mx/09 genomes by introducing Ca/09 HA with two naturally occurring variants expressing S186/I194 (HA-S/I) and P186/L194 (HA-P/L), respectively. The Ca/09 HA with the combination of P186/L194 with either the Ca/09 or Mx/09 backbone resulted in higher and prolonged viral shedding in naïve pigs. This efficiency appeared to be more likely through an advantage in cell surface attachment rather than replication efficiency. Although these mutations occurred within the receptor-binding pocket and the Sb antigenic site, they did not affect serological cross-reactivity. Relative increases of P186 in publicly available sequences from swine H1N1pdm09 viruses supported the experimental data, indicating this amino acid substitution conferred an advantage in swine.

Feature selection methods for identifying genetic determinants of host species in RNA viruses

Despite environmental, social and ecological dependencies, emergence of zoonotic viruses in human populations is clearly also affected by genetic factors which determine cross-species transmission potential. RNA viruses pose an interesting case study given their mutation rates are orders of magnitude higher than any other pathogen – as reflected by the recent emergence of SARS and Influenza for example. Here, we show how feature selection techniques can be used to reliably classify viral sequences by host species, and to identify the crucial minority of host-specific sites in pathogen genomic data. The variability in alleles at those sites can be translated into prediction probabilities that a particular pathogen isolate is adapted to a given host. We illustrate the power of these methods by: 1) identifying the sites explaining SARS coronavirus differences between human, bat and palm civet samples; 2) showing how cross species jumps of rabies virus among bat populations can be readily identified; and 3) de novo identification of likely functional influenza host discriminant markers.

Moving away from genome scan methods used for human GWAS (ultimately inappropriate for the short highly polymorphic genomes of RNA viruses), our work shows the power and potential of multi-class machine learning algorithms in inferring the functional genetic changes associated with phenotypic change (e.g. crossing a species barrier). We show that even distantly related viruses within a viral family share highly conserved genetic signatures of host specificity; reinforce how fitness landscapes of host adaptation are shaped by host phylogeny; and highlight the evolutionary trajectories of RNA viruses in rapid expansion and under great evolutionary pressure. We do so by (for each dataset) unveiling a set of phenotype characteristic mutations which are shown to be functionally relevant, thus providing new insights into phenotypic relationships between RNA viruses. These methods also provide a solid statistical framework with which the degree of host adaptation can be inferred, thus serving as a valuable tool for studying host transition events with particular relevance for emerging infectious diseases. These methods can then serve as rigorous tools of emergence potential assessment, specifically in scenarios where rapid host classification of newly emerging viruses can be more important than identifying putative functional sites.

Characterization of a porcine intestinal epithelial cell line for influenza virus production

We have developed a porcine intestine epithelial cell line, designated SD-PJEC for the propagation of influenza viruses. The SD-PJEC cell line is a subclone of the IPEC-J2 cell line, which was originally derived from newborn piglet jejunum. Our results demonstrate that SD-PJEC is a cell line of epithelial origin that preferentially expresses receptors of oligosaccharides with Sia2-6Gal modification. This cell line is permissive to infection with human and swine influenza A viruses and some avian influenza viruses, but poorly support the growth of human-origin influenza B viruses. Propagation of swine-origin influenza viruses in these cells results in a rapid growth rate within the first 24 h post-infection and the titres ranged from 4 to 8 log(10) TCID(50) ml(-1). The SD-PJEC cell line was further tested as a potential alternative cell line to Madin-Darby canine kidney (MDCK) cells in conjunction with 293T cells for rescue of swine-origin influenza viruses using the reverse genetics system. The recombinant viruses A/swine/North Carolina/18161/02 (H1N1) and A/swine/Texas/4199-2/98 (H3N2) were rescued with virus titres of 7 and 8.25 log(10) TCID(50) ml(-1), respectively. The availability of this swine-specific cell line represents a more relevant substrate for studies and growth of swine-origin influenza viruses.

Mutations in the GM1 binding site of simian virus 40 VP1 alter receptor usage and cell tropism

Polyomaviruses are nonenveloped viruses with capsids composed primarily of 72 pentamers of the viral VP1 protein, which forms the outer shell of the capsid and binds to cell surface oligosaccharide receptors. Highly conserved VP1 proteins from closely related polyomaviruses recognize different oligosaccharides. To determine whether amino acid changes restricted to the oligosaccharide binding site are sufficient to determine receptor specificity and how changes in receptor usage affect tropism, we studied the primate polyomavirus simian virus 40 (SV40), which uses the ganglioside GM1 as a receptor that mediates cell binding and entry. Here, we used two sequential genetic screens to isolate and characterize viable SV40 mutants with mutations in the VP1 GM1 binding site. Two of these mutants were completely resistant to GM1 neutralization, were no longer stimulated by incorporation of GM1 into cell membranes, and were unable to bind to GM1 on the cell surface. In addition, these mutant viruses displayed an infection defect in monkey cells with high levels of cell surface GM1. Interestingly, one mutant infected cells with low cell surface GM1 more efficiently than wild-type virus, apparently by utilizing a different ganglioside receptor. Our results indicate that a small number of mutations in the GM1 binding site are sufficient to alter ganglioside usage and change tropism, and they suggest that VP1 divergence is driven primarily by a requirement to accommodate specific receptors. In addition, our results suggest that GM1 binding is required for vacuole formation in permissive monkey CV-1 cells. Further study of these mutants will provide new insight into polyomavirus entry, pathogenesis, and evolution.

Two years after pandemic influenza A/2009/H1N1: what have we learned?

The world had been anticipating another influenza pandemic since the last one in 1968. The pandemic influenza A H1N1 2009 virus (A/2009/H1N1) finally arrived, causing the first pandemic influenza of the new millennium, which has affected over 214 countries and caused over 18,449 deaths. Because of the persistent threat from the A/H5N1 virus since 1997 and the outbreak of the severe acute respiratory syndrome (SARS) coronavirus in 2003, medical and scientific communities have been more prepared in mindset and infrastructure. This preparedness has allowed for rapid and effective research on the epidemiological, clinical, pathological, immunological, virological, and other basic scientific aspects of the disease, with impacts on its control. A PubMed search using the keywords "pandemic influenza virus H1N1 2009" yielded over 2,500 publications, which markedly exceeded the number published on previous pandemics. Only representative works with relevance to clinical microbiology and infectious diseases are reviewed in this article. A significant increase in the understanding of this virus and the disease within such a short amount of time has allowed for the timely development of diagnostic tests, treatments, and preventive measures. These findings could prove useful for future randomized controlled clinical trials and the epidemiological control of future pandemics.