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

Sequence-based detection of emerging antigenically novel influenza A viruses

The detection of evolutionary transitions in influenza A (H3N2) viruses' antigenicity is a major obstacle to effective vaccine design and development. In this study, we describe Novel Influenza Virus A Detector (NIAViD), an unsupervised machine learning tool, adept at identifying these transitions, using the HA1 sequence and associated physico-chemical properties. NIAViD performed with 88.9% (95% CI, 56.5-98.0%) and 72.7% (95% CI, 43.4-90.3%) sensitivity in training and validation, respectively, outperforming the uncalibrated null model-33.3% (95% CI, 12.1-64.6%) and does not require potentially biased, time-consuming and costly laboratory assays. The pivotal role of the Boman's index, indicative of the virus's cell surface binding potential, is underscored, enhancing the precision of detecting antigenic transitions. NIAViD's efficacy is not only in identifying influenza isolates that belong to novel antigenic clusters, but also in pinpointing potential sites driving significant antigenic changes, without the reliance on explicit modelling of haemagglutinin inhibition titres. We believe this approach holds promise to augment existing surveillance networks, offering timely insights for the development of updated, effective influenza vaccines. Consequently, NIAViD, in conjunction with other resources, could be used to support surveillance efforts and inform the development of updated influenza vaccines.

Seasonal antigenic prediction of influenza A H3N2 using machine learning

Antigenic characterization of circulating influenza A virus (IAV) isolates is routinely assessed by using the hemagglutination inhibition (HI) assays for surveillance purposes. It is also used to determine the need for annual influenza vaccine updates as well as for pandemic preparedness. Performing antigenic characterization of IAV on a global scale is confronted with high costs, animal availability, and other practical challenges. Here we present a machine learning model that accurately predicts (normalized) outputs of HI assays involving circulating human IAV H3N2 viruses, using their hemagglutinin subunit 1 (HA1) sequences and associated metadata. Each season, the model learns an updated nonlinear mapping of genetic to antigenic changes using data from past seasons only. The model accurately distinguishes antigenic variants from non-variants and adaptively characterizes seasonal dynamics of HA1 sites having the strongest influence on antigenic change. Antigenic predictions produced by the model can aid influenza surveillance, public health management, and vaccine strain selection activities.

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.

Novel graph-based machine-learning technique for viral infectious diseases: application to influenza and hepatitis diseases

BACKGROUND

Most infectious diseases are caused by viruses, fungi, bacteria and parasites. Their ability to easily infect humans and trigger large-scale epidemics makes them a public health concern. Methods for early detection of these diseases have been developed; however, they are hindered by the absence of a unified, interoperable and reusable model. This study seeks to create a holistic and real-time model for swift, preliminary detection of infectious diseases using symptoms and additional clinical data.

MATERIALS AND METHODS

In this study, we present a medical knowledge graph (MKG) that leverages multiple data sources to analyse connections between different nodes. Medical ontologies were used to enhance the MKG. We applied various graph algorithms to extract key features. The performance of multiple machine-learning (ML) techniques for influenza and hepatitis detection was assessed, selecting multi-layer perceptron (MLP) and random forest (RF) models due to their superior outcomes. The hyperparameters of both graph-based ML models were automatically fine-tuned.

RESULTS

Both the graph-based MLP and RF models showcased the least loss and error rates, along with the most specific, accurate recall, precision and F1 scores. Their Matthews correlation coefficients were also optimal. When compared with existing ML techniques and findings from the literature, these graph-based ML models manifested superior detection accuracy.

CONCLUSIONS

The graph-based MLP and RF models effectively diagnosed influenza and hepatitis, respectively. This underlines the potential of graph data science in enhancing ML model performance and uncovering concealed relationships in the MKG.

A gradient boosting tree model for multi-department venous thromboembolism risk assessment with imbalanced data

Venous thromboembolism (VTE) is the world's third most common cause of vascular mortality and a serious complication from multiple departments. Risk assessment of VTE guides clinical intervention in time and is of great importance to in-hospital patients. Traditional VTE risk assessment methods based on scaling tools, which always require rules carefully designed by human experts, are difficult to apply to large-population scenarios since the manually designed rules are not guaranteed to be accurate to all populations. In contrast, with the development of the electronic health record (EHR) datasets, data-driven machine-learning-based risk assessment methods have proven superior predictability in many studies in recent years. This paper uses the gradient boosting tree model to study the VTE risk assessment problem with multi-department data. There exist two distinct characteristics of VTE data collected at the level of the entire hospital: its wide distribution and heterogeneity across multiple departments. To this end, we consider the prediction task over multiple departments as a multi-task learning process, and introduce the algorithm of a task-aware tree-based method TSGB to tackle the multi-task prediction problem. Although the introduction of multi-task learning improves overall across-department performance, we reveal the problem of task-wise performance decline while dealing with imbalanced VTE data volume. According to the analysis, we finally propose two variants of TSGB to alleviate the problems and further boost the prediction performance. Compared with state-of-the-art rule-based and multi-task tree-based methods, the experimental results show the proposed methods not only improve the overall across-department AUC performance effectively, but also ensure the improvement of performance over every single department prediction.

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.

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.

Measuring Site-specific Glycosylation Similarity between Influenza a Virus Variants with Statistical Certainty

Influenza A virus (IAV) mutates rapidly, resulting in antigenic drift and poor year-to-year vaccine effectiveness. One challenge in designing effective vaccines is that genetic mutations frequently cause amino acid variations in IAV envelope protein hemagglutinin (HA) that create new N-glycosylation sequons; resulting N-glycans cause antigenic shielding, allowing viral escape from adaptive immune responses. Vaccine candidate strain selection currently involves correlating antigenicity with HA protein sequence among circulating strains, but quantitative comparison of site-specific glycosylation information may likely improve the ability to design vaccines with broader effectiveness against evolving strains. However, there is poor understanding of the influence of glycosylation on immunodominance, antigenicity, and immunogenicity of HA, and there are no well-tested methods for comparing glycosylation similarity among virus samples. Here, we present a method for statistically rigorous quantification of similarity between two related virus strains that considers the presence and abundance of glycopeptide glycoforms. We demonstrate the strength of our approach by determining that there was a quantifiable difference in glycosylation at the protein level between WT IAV HA from A/Switzerland/9715293/2013 (SWZ13) and a mutant strain of SWZ13, even though no N-glycosylation sequons were changed. We determined site-specifically that WT and mutant HA have varying similarity at the glycosylation sites of the head domain, reflecting competing pressures to evade host immune response while retaining viral fitness. To our knowledge, our results are the first to quantify changes in glycosylation state that occur in related proteins of considerable glycan heterogeneity. Our results provide a method for understanding how changes in glycosylation state are correlated with variations in protein sequence, which is necessary for improving IAV vaccine strain selection. Understanding glycosylation will be especially important as we find new expression vectors for vaccine production, as glycosylation state depends greatly on the host species.

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.