H3N2 influenza virus antigenicity dynamic evolution analysis method and device

Abstract

The invention discloses a dynamic evolution analysis method and device for H3N2 influenza virus antigenicity, which are used for solving the problems of poor prediction accuracy, large limitation and difficulty in accurately identifying antigen clusters and evolution rules thereof in the current H3N2 influenza virus antigen analysis method. The method comprises the following steps: acquiring a plurality of neuraminidase virus pairs and a plurality of effective antigen characteristics of the H3N2 influenza virus; performing feature quantification on the plurality of neuraminidase virus pairs based on the plurality of effective antigen features, and constructing an antigen relationship prediction data set; performing model training on a pre-constructed double-layer prediction model according to the antigen relation prediction data set to obtain an antigen similarity prediction model; inputting to-be-detected neuraminidase sequence data into the antigen similarity prediction model for antigen relation prediction; and constructing an antigen correlation network based on the antigen relationship prediction result, and carrying out network clustering on the antigen correlation network to obtain an antigen evolution analysis result of the H3N2 influenza virus.

Description

Technical Field

[0001] The present invention relates to the technical field of influenza virus analysis, and in particular to a method for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus, a device for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus, an electronic device and a storage medium. Background Art

[0002] As the main surface proteins of H3N2 influenza virus, hemagglutinin (HA) and neuraminidase (NA) play an important role in the process of antigenic drift. At present, the design of most influenza vaccines is mainly based on HA antigenicity. Although it is effective in inhibiting the spread of influenza, the vaccine has a low protective effect against A / H3N2. Studies have shown that NA antibodies can provide cross-protection to a certain extent and play an important role in preventing influenza virus infection, inhibiting virus spread, alleviating clinical symptoms and enhancing vaccine efficacy. The antigenic evolution of H3N2 influenza virus has an important impact on the selection of vaccine strains. By analyzing the antigenic evolution of H3N2, it can provide a reference for the analysis of influenza virus antigenic evolution and future vaccine development. Therefore, incorporating NA antigenicity into the vaccine development process and exploring the law of its antigenic evolution has become an important research direction to improve the effectiveness of vaccines.

[0003] In order to better understand the antigenicity of H3N2 NA, current studies usually use experimental methods (such as neuraminidase inhibition test) to detect antigenic differences between different NA strains and divide antigenic clusters based on experimental data. However, these traditional methods are complicated to operate, time-consuming, and difficult to obtain results quickly. At the same time, the scattered sources and inconsistent quality of samples further limit the efficiency and applicability of the experiment. In addition, current experimental methods can usually only analyze the antigenicity between limited virus strains, which is difficult to meet the needs of large-scale antigenic analysis. As a result, the results of antigenic analysis are limited and cannot fully reflect the antigenic evolution of NA, which seriously affects the accuracy and comprehensiveness of vaccine strain recommendations.

[0004] The current analytical methods also lack systematic and model-based predictive capabilities, which limits the model's ability to accurately predict viral antigenic relationships. At the same time, due to insufficient training data, existing models are prone to overfitting or prediction bias and cannot effectively capture the complex laws of NA antigenic evolution. There is also a lack of systematic analysis of the antigenic relationship of viral strains. As a result, it is impossible to accurately identify antigen clusters and their evolution laws, making it difficult to provide a scientific basis for vaccine strain recommendations. Summary of the invention

[0005] The present invention provides a method for analyzing the dynamic evolution of H3N2 influenza virus antigenicity, a device for analyzing the dynamic evolution of H3N2 influenza virus antigenicity, an electronic device and a storage medium, which are used to solve or partially solve the technical problems existing in the current H3N2 influenza virus antigen analysis method, such as poor prediction accuracy, large limitations and difficulty in accurately identifying antigen clusters and their evolution laws.

[0006] The present invention provides a method for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus, comprising:

[0007] Obtain multiple neuraminidase virus pairs and multiple effective antigenic features of H3N2 influenza virus;

[0008] quantifying the features of the plurality of neuraminidase virus pairs based on the plurality of effective antigen features to construct an antigen relationship prediction data set;

[0009] Performing model training on a pre-constructed two-layer prediction model according to the antigen relationship prediction data set to obtain an antigen similarity prediction model;

[0010] Inputting the neuraminidase sequence data to be tested into the antigen similarity prediction model to perform antigen relationship prediction;

[0011] An antigen correlation network was constructed based on the antigen relationship prediction results, and network clustering was performed on the antigen correlation network to obtain the antigen evolution analysis results of the H3N2 influenza virus.

[0012] The present invention also provides a device for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus, comprising:

[0013] A data acquisition unit, used to acquire multiple neuraminidase virus pairs and multiple effective antigen features of H3N2 influenza virus;

[0014] A feature quantification unit, configured to perform feature quantification on the plurality of neuraminidase virus pairs based on the plurality of effective antigen features, and construct an antigen relationship prediction data set;

[0015] A model training unit, used to perform model training on a pre-built two-layer prediction model according to the antigen relationship prediction data set to obtain an antigen similarity prediction model;

[0016] An antigen relationship prediction unit, used for inputting the sequence data of the neuraminidase to be tested into the antigen similarity prediction model to perform antigen relationship prediction;

[0017] The antigen evolution analysis unit is used to construct an antigen correlation network based on the antigen relationship prediction result, and perform network clustering on the antigen correlation network to obtain the antigen evolution analysis result of the H3N2 influenza virus.

[0018] The present invention also provides an electronic device, the device comprising a processor and a memory:

[0019] The memory is used to store program code and transmit the program code to the processor;

[0020] The processor is used to execute the method for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus as described above according to the instructions in the program code.

[0021] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium is used to store program code, and the program code is used to execute the method for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus as described in any one of the above items.

[0022] It can be seen from the above technical solutions that the present invention has the following advantages:

[0023] A method for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus is provided. For model construction, multiple neuraminidase virus pairs and multiple effective antigen features of H3N2 influenza virus are first obtained; then, based on the multiple effective antigen features, the multiple neuraminidase virus pairs are characterized and quantified to construct an antigenic relationship prediction data set; then, the pre-constructed double-layer prediction model is trained according to the antigenic relationship prediction data set to obtain an antigenic similarity prediction model. Thus, by fully integrating the multiple effective antigenic features of neuraminidase, a high-precision antigenic relationship prediction model based on machine learning is developed based on the integrated learning method of the double-layer model, so as to achieve high-precision prediction of the antigenic relationship between virus strains in the subsequent process. For the actual analysis process of the antigenicity of H3N2 influenza virus, the neuraminidase sequence data to be tested is input into the antigenic similarity prediction model for antigenic relationship prediction; based on the antigenic relationship prediction results, an antigenic correlation network is constructed, and the antigenic correlation network is clustered to obtain the antigenic evolution analysis results of H3N2 influenza virus. Based on the prediction results, we can further construct an antigen correlation network covering global H3N2 NA data. Through dynamic network and cluster analysis, it will help to quickly identify possible new epidemic strains when viral antigens mutate, reveal the distribution and evolution of antigen clusters, and provide new scientific basis for influenza monitoring and vaccine selection. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0025] Figure 1 A flowchart of a method for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus;

[0026] Figure 2 A schematic diagram of an antigen epitope region and an enzyme catalytic site;

[0027] Figure 3 This is a schematic diagram of the comparison of test set results when using different antigen similarity prediction models;

[0028] Figure 4 A schematic diagram showing the comparison of cross-validation results when using different antigen similarity prediction models;

[0029] Figure 5 A schematic diagram of an antigen correlation network and phylogenetic tree;

[0030] Figure 6 It is a heat map of intra-cluster and inter-cluster similarity of an antigen cluster;

[0031] Figure 7 This is a graph showing the average cluster size and modularity changing with the expansion parameter when clustering a network;

[0032] Figure 8 A schematic diagram showing the comparison between the spatiotemporal prevalence of H3N2 virus based on neuraminidase and that based on hemagglutinin;

[0033] Fig. 9 The structure diagram of a device for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus. DETAILED DESCRIPTION

[0034] The embodiments of the present invention provide a method for analyzing the dynamic evolution of H3N2 influenza virus antigenicity, a device for analyzing the dynamic evolution of H3N2 influenza virus antigenicity, an electronic device and a storage medium, which are used to solve or partially solve the technical problems existing in the current H3N2 influenza virus antigen analysis method, such as poor prediction accuracy, large limitations and difficulty in accurately identifying antigen clusters and their evolution laws.

[0035] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] As an example, in order to better understand the antigenicity of H3N2 NA, current studies usually use experimental methods (such as neuraminidase inhibition test) to detect antigenic differences between different NA strains and divide antigenic clusters based on experimental data. However, these traditional methods are complicated to operate, time-consuming, and difficult to obtain results quickly. At the same time, the scattered sources and inconsistent quality of samples further limit the efficiency and applicability of the experiment. In addition, current experimental methods can usually only analyze the antigenicity between limited virus strains, which is difficult to meet the needs of large-scale antigenic analysis. As a result, the results of antigenic analysis are limited and cannot fully reflect the antigenic evolution of NA, which seriously affects the accuracy and comprehensiveness of vaccine strain recommendations.

[0037] The current analytical methods also lack systematic and model-based predictive capabilities, which limits the model's ability to accurately predict viral antigenic relationships. At the same time, due to insufficient training data, existing models are prone to overfitting or prediction bias and cannot effectively capture the complex laws of NA antigenic evolution. There is also a lack of systematic analysis of the antigenic relationship of viral strains. As a result, it is impossible to accurately identify antigen clusters and their evolution laws, making it difficult to provide a scientific basis for vaccine strain recommendations.

[0038] Therefore, one of the core invention points of the embodiment of the present invention is to provide a method for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus based on neuraminidase. By fully integrating multiple key features of neuraminidase (such as epitope characteristics, physicochemical properties, glycosylation sites and catalytic sites), based on an integrated learning method of multiple machine learning models, a high-precision antigen relationship prediction model based on machine learning is developed to achieve high-precision prediction of antigenic relationships between virus strains. Based on the prediction results, an antigen correlation network covering global H3N2 NA data is further constructed. Through dynamic network and cluster analysis, it is helpful to quickly identify possible new epidemic strains when viral antigens mutate, reveal the distribution and evolution of antigen clusters, and provide new scientific basis for influenza monitoring and vaccine selection.

[0039] Reference Figure 1 , shows a flow chart of the steps of a method for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus provided by an embodiment of the present invention, which may specifically include the following steps:

[0040] Step 101, obtaining multiple neuraminidase virus pairs and multiple effective antigen features of H3N2 influenza virus;

[0041] In some embodiments, the process of obtaining multiple neuraminidase virus pairs and multiple effective antigen features of H3N2 influenza virus can be achieved by performing the following sub-steps S01 to S03:

[0042] Step S01: collecting neuraminidase sequence data and neuraminidase inhibition test data of H3N2 influenza virus;

[0043] First, the present invention collected all H3N2 subtype influenza virus neuraminidase sequence data from the Global Initiative on Sharing All Influenza Data (GISAID) database, which covered virus samples from different months and continents between 1968 and 2024.

[0044] In addition, the present invention also collects extensive literature information and screens out test data pairs with both homologous titers and heterologous titers. A total of 377 pairs of neuraminidase inhibition (NAI) test data were sorted out. These data come from neuraminidase inhibition tests between different virus strains, and record the antigenic differences between virus pairs. Among them, the neuraminidase inhibition test data provide experimental support for the antigenic relationship of viral neuraminidase, which plays an important role in calculating antigenic distance and training antigen similarity prediction models. These data contain rich antigenic relationship information, which, combined with subsequent rigorous screening and optimization, can improve the reliability and availability of subsequent analysis.

[0045] Step S02: performing double optimization of sequence screening on the neuraminidase sequence data to obtain optimized sequence data, and performing feature selection based on the optimized sequence data to extract multiple effective antigen features;

[0046] In order to improve the data quality and the effect of model training, it is necessary to perform multi-step data processing on the collected neuraminidase sequence data and neuraminidase inhibition test data.

[0047] In some embodiments, the process of performing sequence screening and double optimization on the neuraminidase sequence data to obtain the optimized sequence data can be achieved by executing the following sub-steps S11 to S17:

[0048] Step S11: uniformly distribute and randomly sample the neuraminidase sequence data to obtain sampled neuraminidase sequence data;

[0049] The processing of neuraminidase sequence data mainly involves eliminating low-quality or incomplete sequences and using a unified naming format (such as virus strain name, date, and place of origin) for standardized management.

[0050] For the sequence sampling process, firstly, random sampling is performed from all neuraminidase sequence data, and influenza samples from different continents (North America, Europe, Asia, South America, Africa, Oceania) between 1968 and 2024 are selected. The process uses the month as the time point for evenly distributed sampling to ensure global representativeness. Specifically, for each sampling node of each month in each continent, 7 sequences can be randomly selected from the sequence data corresponding to that month. When there are less than 7 sequence data for that month, all sequences for that month are sampled.

[0051] The adoption of the above-mentioned equal random sampling strategy can make the data well balanced in terms of geographical distribution and time dimension. After uniformly distributed random sampling, the amount of data in each continent is roughly the same, which can effectively solve the bias of model training in regions with high data volume (such as North America and Europe), so that the model can fully learn the antigenic characteristics of regions with low data volume. In addition, by uniformly extracting samples every month, the data can be continuous and representative in time. It helps to reflect the trends and laws of H3N2 antigenicity in the long-term evolution process. Through uniformly distributed random sampling, the repetitiveness and redundancy of samples are reduced, making the training data more representative and diverse. Therefore, by selecting sample data from different periods and geographical regions, the universality and wide applicability of the subsequent construction of the prediction model can be greatly improved, and the antigenic changes of the H3N2 virus in the neuraminidase level can be accurately reflected.

[0052] Although the possibility of occurrence is small, when sampling sequences globally, there may be cases where the number of sequences in some months or regions is small or even zero. When the number of sequences is 0, the use of methods such as duplication of data for supplementation may introduce unreal noise or bias, affecting the accuracy of antigen analysis. Therefore, there is no need to duplicate or supplement the data in this case. It is understandable that although the sequences in some months may be insufficient, this local missing situation has little impact on the overall analysis. The representativeness of data on a global scale can be supplemented and balanced by data from other months and continents, thereby reducing the impact of local deficiencies.

[0053] To further verify the reliability of the sampling strategy, the embodiment of the present invention also tried a percentage-based proportional sampling method. That is, 5% of the sequences (at least 1) were selected each month and each continent. Subsequent analysis results showed that the two sequence sampling methods were highly consistent in the results of antigen cluster division and vaccine strain matching. It can be seen that the uniform sampling method can better avoid the bias caused by changes in the amount of sample data.

[0054] Step S12: performing multiple sequence alignment on the sampled neuraminidase sequence data, performing sequence alignment based on the multiple sequence alignment results, and screening out low-quality neuraminidase sequences whose vacancies exceed a preset vacancy threshold during the sequence alignment process;

[0055] The sampled neuraminidase sequence data can be aligned using the MAFFT (Multiple Alignment using Fast Fourier Transform) multiple sequence alignment tool, and sequence alignment can be performed based on the multiple sequence alignment results. Sequence alignment and alignment can make the sequences relatively consistent, eliminating possible sequence errors and insertions and deletions. In addition, the present invention also sets a strict quality filtering standard during the sequence alignment process. That is, low-quality neuraminidase sequences with gaps exceeding 10% are deleted.

[0056] Step S13: selecting at least one target neuraminidase sequence with a vacancy from all retained neuraminidase sequences;

[0057] Subsequently, the Bayesian interpolation method can be used to dynamically fill in a small number of gaps in the retained neuraminidase sequences. Therefore, at least one target neuraminidase sequence with a gap can be selected from all the retained neuraminidase sequences.

[0058] Step S14: for each target neuraminidase sequence, determine the amino acid vacancy site of the target neuraminidase sequence, and record the sequence context amino acids upstream and downstream of the amino acid vacancy site;

[0059] For each target neuraminidase sequence, determine the vacant amino acid site and record the upstream and downstream sequence context amino acids (e.g., 3 amino acids before and after).

[0060] Step S15: screening out, from all the retained complete neuraminidase sequences, at least one target complete neuraminidase sequence having a similarity match degree with the sequence context amino acid higher than a preset match degree threshold;

[0061] Then, you can search for related sequences, that is, sequences with similar sequence context amino acids (3 amino acids before and after) to the position of the gap in the target neuraminidase sequence. Specifically, you can use the complete neuraminidase sequence that does not need to be filled in (which can be understood as other high-quality sequences in the comparison results) as a reference to find the top 10 complete neuraminidase sequences that are most similar to the target neuraminidase sequence context, and use them as reference sequences to dynamically fill the gaps in the target neuraminidase sequence.

[0062] Step S16: Based on at least one target complete neuraminidase sequence, predicting gap filling values ​​of amino acid gaps by Bayesian interpolation, and dynamically filling the target neuraminidase sequence based on the gap filling values ​​to obtain a filled neuraminidase sequence;

[0063] For example, suppose there is a gap __ at a certain position in the target neuraminidase sequence, and its context fragments are ALK and VGT. The top 10 sequences with the best context matches are selected from all the retained complete neuraminidase sequences (high-quality sequences), such as ALKQVGT and ALKSVGT. The most likely value of the gap amino acid is predicted by the Bayesian interpolation method, and the gap is finally filled with "Q". In the set of similar sequences, for the position where the target gap is located, the distribution probability of the amino acid at that position is counted.

[0064] According to Bayes' theorem, the posterior probability of the amino acid at the target position is calculated as follows:

[0065]

[0066] in, ) represents the prior probability of the amino acid in the database; represents the conditional probability of the context given the amino acid X; Represents the overall probability of the context in all sequences.

[0067] When the posterior probability of the amino acid gaps that can be filled corresponding to each target complete neuraminidase sequence is calculated, the amino acid with the highest probability is filled into the target gap to complete the dynamic correction. Assuming that the probability distribution is relatively uniform (i.e., the probabilities of multiple amino acids are close), the mode of the amino acids with the highest probability is selected as the final interpolation result. Dynamic correction can improve data integrity and reduce data loss caused by the elimination of low-quality sequences. Thus, the accuracy of interpolation is enhanced by dynamically considering the context and sequence similarity.

[0068] Step S17: Integrate all filled neuraminidase sequences and all complete neuraminidase sequences, and add the H3N2 recommended vaccine strains of previous years to obtain optimized sequence data.

[0069] All filled neuraminidase sequences and all complete neuraminidase sequences were integrated, and the H3N2 vaccine strains recommended by the World Health Organization over the years were added to these sequence data. Through the above processing steps, 8992 neuraminidase sequences were obtained as optimized sequence data. Thus, through comparison and filtering, the data finally retained were high-quality neuraminidase sequence data, providing a reliable basis for subsequent feature extraction and quantification.

[0070] Extracting effective features from the virus sequence is the key to predicting whether the antigen is similar to a known strain. In the embodiment of the present invention, the antigen features are divided into four categories according to the characteristics of neuraminidase. They are antigen epitope features, physicochemical properties, N-glycosylation site features, and enzyme catalytic site features.

[0071] In some embodiments, the process of performing feature selection based on the optimized sequence data to extract multiple effective antigen features can be achieved by executing the following sub-steps S21 to S27:

[0072] Step S21: using a variety of different prediction tools to perform structural analysis on each neuraminidase sequence in the optimized sequence data, and recording all residues predicted to be antigenic epitopes;

[0073] First, the antigen epitope features are extracted. In the embodiment of the present invention, based on the 7U4E template in the PDB (Protein Data Bank) database, the structure of H3N2 neuraminidase is analyzed and its potential structural epitopes are predicted by using four tools, namely Discotope-2.0, Discotope-3.0, SEMA2.0, and ScanNet, to improve the accuracy and reliability of the prediction.

[0074] These four tools each have their own unique prediction principles and algorithm methods, thus providing different perspectives in antigen epitope recognition. The present invention combines the prediction results of these tools to propose an optimized multi-tool comprehensive prediction strategy, overcoming the limitations of a single prediction method.

[0075] Among them, the main prediction principle of Discotope-2.0 is based on the solvent accessibility (SASA) of protein residues and the statistical tendency of amino acids. It assumes that antibodies mainly bind to areas exposed on the protein surface, and further combines the distribution tendency of amino acids in known epitope data to calculate the epitope score of each residue.

[0076] DiscoTope-3.0 is a structure-based epitope prediction tool. It uses high-dimensional structural representations generated by protein inverse folding models to predict B cell conformational epitopes. Unlike traditional methods that rely only on solvent accessibility (SASA) and geometric properties, DiscoTope-3.0 converts the three-dimensional structure of proteins into feature vectors that can be processed by deep learning models. These include surface accessibility of residues, AlphaFold structural quality scores (predicted Local Distance DifferenceTest, pLDDT), and sequence length correction scores.

[0077] SEMA2.0 is a multimodal deep learning tool that combines protein sequence and three-dimensional structure information. By introducing a large-scale pre-trained language model (Evolutionary Scale Modeling 2, ESM2) and a structure-aware protein module (SaProt) combined with geometric patterns, SEMA2.0 is able to capture epitope characteristics from the dual dimensions of sequence and structure.

[0078] ScanNet is a geometric deep learning model. ScanNet extracts the geometric and physicochemical features of binding sites from the three-dimensional structure of proteins, and uses convolutional neural networks (CNNs) to perform multi-scale modeling of residue spatial characteristics through the geometric relationship of residue neighborhoods and chemical information such as solvent accessibility and charge distribution to predict and recognize antigen conformational epitopes.

[0079] Step S22: Calculate the number of times each residue is predicted to be an antigenic epitope, and take the residues whose antigenic epitope prediction times are greater than or equal to a preset times threshold as potential structural epitopes;

[0080] In order to integrate the prediction results of the above four tools, a weighted comprehensive strategy is designed in the embodiment of the present invention. During the data processing process, the prediction results of each tool are sorted and counted, and the residue numbers predicted as antigenic epitopes by each tool are recorded. Then, the number of times predicted as an epitope by the four tools is calculated for each residue. For example, if a residue is predicted as an epitope by two tools, the number is recorded as 2. A number threshold (such as twice) can be set in advance, that is, the residues predicted as epitopes by at least two tools are considered to have higher reliability, and these residues can be screened out as weighted epitope prediction results.

[0081] Specifically, for each residue, if the number of predictions in the four tools reaches or exceeds two times, the residue is considered to be a possible antigenic epitope, that is, a potential structural epitope. The weighted comprehensive strategy combines the advantages of the four tools, retaining the robustness of DiscoTope-3.0 in predicting structural models, and making full use of ScanNet's modeling capabilities for geometric details and SEMA2.0's multimodal adaptability. In this way, the final prediction results can effectively reduce the risk of false positives or false negatives that may be generated by a single tool, and significantly improve the accuracy and reliability of the prediction.

[0082] In the embodiment of the present invention, selecting twice as the threshold can not only achieve a good balance between sensitivity and specificity, but also cover most potential real epitopes while avoiding omissions caused by too high standards. In general, selecting twice as the threshold can meet actual needs.

[0083] Step S23: performing cluster analysis based on all potential structural epitopes, determining the number of clusters according to the Euclidean distance relationship and silhouette coefficient between each potential structural epitope, and obtaining five antigen epitope features;

[0084] The predicted potential structural epitopes were clustered using the K-means clustering algorithm, and the number of clusters was determined based on the Euclidean distance relationship and the silhouette score between the epitopes. The embodiment of the present invention finally determined five antigenic epitopes through clustering. They were named N2_A, N2_B, N2_C, N2_D and N2_E. The antigenic epitope region is as follows Figure 2 As shown. These epitopes are derived by analyzing the frequency distribution of viral mutation positions, ensuring the accuracy and applicability of epitope definition. At the same time, the extraction of epitope features provides important input features for subsequent antigenicity prediction.

[0085] The site information contained in each antigen epitope is shown in Table 1 below:

[0086]

[0087] Table 1: Site information contained in each antigen epitope

[0088] Step S24: screening hydrophobic residues, charge distribution residues, charge polarity residues, molecular volume residues and solvent accessible surface area residues from each neuraminidase sequence in the optimized sequence data;

[0089] The present invention screened based on biological significance and selected five physicochemical properties that play an important role in antigen-antibody interaction, including hydrophobicity, charge, polarity, volume and (solvent) accessible surface area.

[0090] Hydrophobicity is one of the key driving forces for antigen-antibody binding. Antibodies tend to bind to regions with lower hydrophobicity. These regions are usually located on the surface of antigens, which facilitates antibody recognition. At the same time, aromatic hydrophobic residues (such as tyrosine, phenylalanine, and tryptophan) play an important role in the antigen-antibody interface. These residues not only enhance binding affinity through π-π interactions, but also provide stable binding sites for antigen epitopes through their unique geometry.

[0091] Charge distribution directly affects the binding specificity of antibodies to antigens. The charge distribution in the epitope region of an antigen usually contains a reasonable combination of positive and negative charged residues. This distribution enhances the stability of the binding through electrostatic interactions.

[0092] The binding site of an antibody is usually rich in negatively charged residues (such as aspartic acid and glutamic acid). The antigen epitope is rich in positively charged residues (such as arginine and lysine). This charge complementarity phenomenon not only enhances the binding strength, but also optimizes the geometric matching of the binding interface. The epitope region is usually exposed to the solvent and has a high accessible surface area. The high exposure of these regions allows antibodies to more easily contact and bind to the epitope.

[0093] In addition, the solvent accessible surface area is also closely related to the conformational flexibility of the antigen epitope, which increases the adaptability of the antibody to the epitope. Polar residues (such as glutamine, asparagine, serine, and threonine) significantly increase the affinity and stability of antigen-antibody binding by forming weak interactions such as hydrogen bonds. These residues tend to be distributed on the surface of the antigen, further enhancing the specificity of binding.

[0094] The molecular volume features reflect the geometric contribution of the residues. Especially in complex antigen epitopes, the bulkier residues (such as tryptophan and phenylalanine) directly promote the antigen-antibody interaction by providing more binding surface. At the same time, the small volume residues (such as glycine and alanine) indirectly optimize the binding interface by providing conformational flexibility. These geometric properties enable the antigen epitope to better adapt to the antibody binding site.

[0095] Step S25: performing feature selection using a random forest model based on all hydrophobic residues, charge distribution residues, charge polarity residues, molecular volume residues, and solvent accessible surface area residues, and screening out five physicochemical property features representing hydrophobicity, charge, polarity, volume, and accessible surface area, respectively, based on feature importance scores;

[0096] Each physicochemical property often has multiple quantitative representation methods in bioinformatics databases. In order to screen out the specific parameter values ​​that best represent each physicochemical property, the quantitative values ​​of the five physicochemical property characteristics of H3N2 neuraminidase were obtained from the AAindex database (Amino Acid Index Database). In the embodiment of the present invention, a random forest model was used for feature selection to construct a feature matrix.

[0097] Among them, each row of the feature matrix corresponds to each residue of H3N2 neuraminidase, and each column is the value of a certain parameter. Each residue is labeled as an antigenic epitope (positive sample, 1) or a non-epitope (negative sample, 0) based on the existing experimental data. During the training process of the random forest model, multiple decision trees are used to classify whether the residue is an antigenic epitope, and the contribution of each parameter to the prediction task is evaluated, and the feature importance score is generated according to the information gain. Through the feature selection and screening of random forest, the representative features finally selected include BLAS910101 (hydrophobicity), FAUJ880111 (charge), RADA880108 (polarity), GRAR740103 (volume) and ROSG850101 (accessible surface area).

[0098] It is understandable that the purpose of using the random forest model in this step is to determine whether each residue is an antigenic epitope and to find out which physicochemical property values ​​are most important for this judgment. In other words, each physicochemical property is scored according to "which physicochemical properties are most helpful in correctly classifying the antigenic epitope", which is called the feature importance score. The predicted result of the antigenic epitope is used as the classification label, and the physicochemical property value is used as the input.

[0099] Step S26: predicting the N-glycosylation site of each neuraminidase sequence in the optimized sequence data to obtain the N-glycosylation site characteristics;

[0100] The N-glycosylation site characteristics can be obtained by predicting the N-glycosylation sites of the neuraminidase sequence using the NetNGlyc website prediction tool.

[0101] Step S27: Based on the data collection and analysis, the enzyme catalytic site characteristics of the H3N2 influenza virus neuraminidase are determined.

[0102] Based on the literature, the present invention collects 13 H3N2 neuraminidase catalytic sites as enzyme catalytic site characteristics, the site schematic diagram is as follows Figure 2 shown.

[0103] It should be pointed out that the 12 features selected by the present invention have been verified and can significantly improve the accuracy and robustness of the prediction based on the H3N2 neuraminidase-antigen relationship. In addition, other feature selection methods can be used to replace the random forest model for the selection of physical and chemical property features. For example, automated feature selection algorithms (SelectKBest (a feature selection method based on univariate statistical tests), recursive feature elimination) can quickly identify important features through statistical analysis or recursive screening. For another example, model-based feature selection (such as LASSO regression (Least Absolute Shrinkage and Selection Operator, a linear regression analysis method)) can effectively cope with high-dimensional data and avoid overfitting. Alternatively, feature extraction methods based on deep learning (such as convolutional neural networks) can improve the level of automation of feature selection to further improve the accuracy and versatility of the model. It is understandable that the present invention is not limited to this.

[0104] Step S03: The neuraminidase inhibition test data is first deduplicated, and then antigen distance quantification is performed to obtain multiple neuraminidase virus pairs.

[0105] The deduplication method for the neuraminidase inhibition test data is as follows: for multiple neuraminidase inhibition test results for the same pair of H3N2 viruses, the median is taken as the final neuraminidase inhibition test titer value, and redundant data is removed to ensure data consistency. This step not only reduces the redundancy in the experimental data, but also eliminates the interference of repeated experimental data on the model training results.

[0106] The present invention obtains 377 pairs of neuraminidase virus data through the above data processing of the neuraminidase inhibition test data.

[0107] In order to effectively train the model later, it is necessary to quantify the antigenic differences between the 377 virus pairs collected by calculating the antigenic distance to obtain the 377 neuraminidase virus pairs used for model training. The calculation formula of the antigenic distance is as follows:

[0108]

[0109] Specifically, represents the NA antigenic distance between strain a and strain b; To use antiserum against virus b Detect the NAI titer (homologous titer) of virus b; To use antiserum against virus b The NAI titer (heterologous titer) of virus a was tested. If the absolute value of the NA antigenic distance between a pair of strains was greater than or equal to 2, they were classified as an antigenically dissimilar pair; otherwise, the pair of strains was considered an antigenically similar pair.

[0110] Step 102, quantifying the characteristics of the plurality of neuraminidase virus pairs based on the plurality of effective antigen characteristics, and constructing an antigen relationship prediction data set;

[0111] After obtaining 12 effective antigen features in four categories through the previous steps, it is necessary to quantify the neuraminidase virus pairs in combination with these effective antigen features so that the subsequent model can perform machine learning training and antigen similarity prediction.

[0112] In some embodiments, the process of quantifying the features of multiple neuraminidase virus pairs based on multiple effective antigen features and constructing an antigen relationship prediction data set can be achieved by executing the following sub-steps S31 to S35:

[0113] Step S31: for each antigenic epitope feature, comparing whether the amino acids of each site in the corresponding epitope region of the neuraminidase virus pair are different, and taking the number of differences as the antigenic epitope feature value of the neuraminidase virus pair in the antigenic epitope feature;

[0114] For each antigenic epitope, each pair of viruses was compared to see if there were any differences in the amino acids at each site in the corresponding epitope region, and the number of differences was used as a characteristic value to reflect the potential for antigenic changes in the virus at that epitope.

[0115] Step S32: For each physicochemical property feature, calculate the difference value of the neuraminidase virus pair at each site, and select the three largest difference values ​​to calculate the average value as the physicochemical property feature value of the neuraminidase virus pair in the physicochemical property feature;

[0116] For each physicochemical property characteristic, the change value of each pair of virus sequences in each physicochemical property is calculated separately. The specific calculation method is: through the 20 amino acid physicochemical characteristic data in the AAindex database, the difference value of each site between the virus pairs is calculated, and the average value of the three largest difference values ​​is taken.

[0117] Step S33: calculating the number of differences in the N-glycosylation sites of the neuraminidase virus pair as the N-glycosylation characteristic value of the N-glycosylation site characteristics of the neuraminidase virus pair;

[0118] The N-glycosylation characteristic value was obtained by calculating the number of differences in N-glycosylation sites between the virus pairs.

[0119] Step S34: Calculate the Euclidean distance value from each differential site to the enzyme catalytic site in the neuraminidase virus pair, and select the three largest Euclidean distance values ​​to calculate the average value as the enzyme catalytic site characteristic value of the neuraminidase virus pair in the enzyme catalytic site characteristic;

[0120] The Euclidean distance from each differential site in the virus pair to the catalytic site was calculated, and the average of the three shortest Euclidean distances was selected as the characteristic value of the enzyme catalytic site.

[0121] Step S35: Integrate the antigenic epitope characteristic values, physicochemical property characteristic values, N-glycosylation characteristic values ​​and enzyme catalytic site characteristic values ​​of all neuraminidase virus pairs to construct an antigenic relationship prediction data set.

[0122] Step 103, performing model training on the pre-built two-layer prediction model according to the antigen relationship prediction data set to obtain an antigen similarity prediction model;

[0123] Traditional related methods mostly select a single machine learning algorithm as the final prediction model. The present invention uses the Stacking ensemble learning method to combine a variety of different machine learning algorithms to construct an antigenicity prediction model. Specifically, based on the classification ability of the comprehensive model, five models were selected: Logistic Regression (Logistic Regression), Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Random Forest (RandomForest, RF) and XGBoost (Extreme Gradient Boosting, a gradient boosting algorithm). In this way, the characteristics and advantages of various machine learning algorithms are fully utilized to build an antigenicity prediction model with better performance.

[0124] More specifically, the two-layer prediction model proposed in the present invention includes a first-layer basic prediction model and a second-layer comprehensive prediction model. Among them, the first-layer basic prediction model includes multiple different prediction sub-models. Logistic regression, support vector machine, K nearest neighbor, random forest and XGBoost can be selected as prediction sub-models respectively. And Logistic regression is selected as the second-layer comprehensive prediction model.

[0125] In some embodiments, the process of training a pre-built two-layer prediction model according to the antigen relationship prediction data set to obtain an antigen similarity prediction model can be achieved by executing the following sub-steps S41 to S45:

[0126] Step S41: dividing the antigen relationship prediction data set into an antigen relationship prediction training set and an antigen relationship prediction test set according to a preset division ratio;

[0127] In actual situations, the number of antigenically similar and different virus pairs in the antigenic relationship prediction dataset is unbalanced. To avoid deviations in the prediction results of the model, the SMOTE (Synthetic Minority Over-sampling Technique) technology can also be used to balance the dataset. SMOTE is used to generate antigenically similar virus pairs with the same number of antigenically dissimilar virus pairs, thereby enhancing data balance. The balanced dataset is divided into an antigenic relationship training set and an antigenic relationship test set in a ratio of 7:3. Among them, the antigenic relationship training set is used for model training, and the antigenic relationship test set is used to verify the generalization performance of the model.

[0128] Step S42: performing model training on each prediction sub-model respectively using the antigen relationship prediction training set;

[0129] Combined with the above introduction, the present invention introduces the Stacking ensemble learning method. The basic model is constructed in the first layer, and Logistic regression, support vector machine, K nearest neighbor, random forest and XGBoost are used as prediction sub-models for model training. Each prediction sub-model learns the antigen relationship prediction training set and generates the corresponding prediction results as features input to the second layer, that is, the two-layer comprehensive prediction model.

[0130] Step S43: inputting the prediction results of each prediction sub-model into the second-layer comprehensive prediction model to perform model training based on comprehensive prediction;

[0131] The two-layer comprehensive prediction model uses a meta-learner and a Logistic regression model with relatively stable performance. During the training process, the final prediction output is generated by integrating the prediction results of each prediction sub-model.

[0132] Step S44: During the model training process, cross-validation is used to perform hierarchical training on the prediction results of each prediction sub-model, and model parameters of the two-layer comprehensive prediction model and each prediction sub-model are optimized by grid parameter adjustment and random parameter adjustment;

[0133] In order to avoid overfitting, during the model training process, cross-validation is used to perform layered training on the prediction results of each prediction sub-model to ensure that the model of each layer can be better generalized. In addition, the hyperparameters of each algorithm can be optimized by grid parameter adjustment method and random parameter adjustment method to further improve the model performance.

[0134] Step S45: using the antigen relationship prediction test set to perform model evaluation on the trained two-layer comprehensive prediction model and each trained prediction sub-model, and output the final antigen similarity prediction model.

[0135] The model evaluation mainly uses accuracy, precision, F1 score, recall rate and AUC (Area Under the Curve) as evaluation indicators, and verifies the actual effect of the model by drawing the ROC (Receiver Operating Characteristic Curve) characteristic curve.

[0136] In order to verify the feasibility of the technical solution provided by the present invention, illustratively, Figure 3 A schematic diagram showing the comparison of test set results when using different antigen similarity prediction models is shown. Figure 4 A schematic diagram showing the comparison of cross-validation results when using different antigen similarity prediction models is shown.

[0137] The Stacking ensemble model proposed in the present invention as an antigen similarity prediction model is superior to a single model in most comprehensive performance indicators. For example, in the independent test set, the accuracy (0.852), F1 score (0.833), recall (0.833) and precision (0.833) of the Stacking ensemble model are all the best. In terms of AUC value, the performance of the Stacking ensemble model (AUC=0.918) is slightly lower than that of the random forest (AUC=0.922). In cross-validation, the accuracy (0.875), F1 score (0.882), recall (0.892) and precision (0.875) of the Stacking ensemble model are all the best. In terms of AUC value, the performance of the Stacking ensemble model (AUC=0.948) is slightly lower than that of the random forest (AUC=0.949). This phenomenon shows that although the random forest has a slight advantage in distinguishing ability, the Stacking ensemble model achieves overall performance optimization in multiple indicators by integrating the prediction results of multiple models. Therefore, compared with the traditional single model method, the Stacking integrated model combines the advantages of multiple algorithms and effectively improves the prediction accuracy of antigen-similar and variant virus pairs through SMOTE data balancing processing.

[0138] It should be pointed out that the present invention constructs an antigen relationship prediction model based on the Stacking ensemble learning model. Those skilled in the art can also use other machine learning methods, such as LightGBM (Light Gradient Boosting Machine, a machine learning algorithm based on a gradient boosting framework), CatBoost (a machine learning algorithm based on a gradient boosting framework), neural network models, etc. These alternatives can achieve similar antigen relationship prediction effects. Thereby further enriching the technical implementation methods and application scenarios of the present invention.

[0139] Among them, the models listed above can be used as alternatives to the Stacking integrated model and used alone for antigen relationship prediction. It can also be used as a prediction sub-model of the basic model in the Stacking integrated model, and used in combination with models such as Logistic regression, support vector machine, K nearest neighbor, random forest and XGBoost. It should be noted that the more sub-models there are in the Stacking integrated model, the more training time and computing resource requirements will increase significantly. In particular, the neural network model has a lower training efficiency than models such as LightGBM, and increased complexity will lead to increased difficulty in parameter adjustment and optimization. Therefore, those skilled in the art can select a suitable model as a first-layer basic model or a second-layer comprehensive prediction model based on actual conditions and comprehensive training requirements. It is understandable that the present invention is not limited to this.

[0140] Step 104, inputting the neuraminidase sequence data to be tested into the antigen similarity prediction model to perform antigen relationship prediction;

[0141] This step mainly predicts the antigenic relationship of the virus sequence data actually required to be detected based on the trained antigenic similarity prediction model. The antigenic relationship prediction result is that the H3N2 neuraminidase virus pair in the neuraminidase inhibition test data is antigenically similar, or antigenically dissimilar.

[0142] In the specific implementation, after the prediction model is obtained through training, the antigenic relationship prediction can be performed by pairing the collected neuraminidase sequence data in pairs to obtain the antigenic similarity probability and dissimilarity probability of the virus pair. The core basis for antigenic similarity prediction is the probability ratio of antigenic changes. That is, the ratio of the similarity probability to the dissimilarity probability (oddsratio) is first calculated, and the logarithm of the comparison value is further taken as the standard for judging the antigenic relationship.

[0143] When the logarithm of the ratio is greater than 0, the viruses are considered to be antigenically similar. When the logarithm of the ratio is less than or equal to 0, the viruses are considered to be antigenically dissimilar. Compared with the traditional method of calculating antigenic relationships, the present invention can more intuitively reflect the significance of antigenic relationships between virus strains through logarithmic conversion of probability ratios, avoiding the inaccuracy that may be caused by a single probability threshold.

[0144] There is no need to rely on complex experimental facilities or consumables. By building a machine learning model, the prediction of antigen similarity can be quickly completed, which reduces experimental time and saves experimental condition costs. It can also be expanded to the analysis of large-scale virus data, significantly improving the efficiency of antigen monitoring.

[0145] Step 105: construct an antigen correlation network based on the antigen relationship prediction result, and perform network clustering on the antigen correlation network to obtain the antigen evolution analysis result of the H3N2 influenza virus.

[0146] In some embodiments, the process of constructing an antigen correlation network based on the antigen relationship prediction results, and performing network clustering on the antigen correlation network to obtain the antigen evolution analysis results of the H3N2 influenza virus can be achieved by executing the following sub-steps S51 to S55:

[0147] Step S51: network visualization of the antigen relationship prediction results;

[0148] The antigen relationship prediction results were visualized using Cytoscape software (an open source biological network visualization and analysis software).

[0149] Step S52: taking each virus strain in the neuraminidase sequence data to be tested as a node, and taking the antigenic similarity relationship between the virus strains as the node edge, and constructing an antigenic correlation network of the H3N2 influenza virus based on each node and each node edge;

[0150] Each H3N2 virus strain is used as a node in the network, and antigenically similar virus pairs are connected by edges to form an Antigen Correlation Network (ACNet). Antigenically similar viruses are clustered together, making it easier to observe and analyze the antigenic relationship between different virus strains.

[0151] Step S53: constructing a phylogenetic tree of H3N2 influenza virus based on the antigen correlation network, wherein the phylogenetic tree is used to reflect the genetic continuity and evolution process between virus strains;

[0152] Through the antigen correlation network, five major antigen clusters can be preliminarily identified and a phylogenetic tree can be constructed. Figure 5 As shown in Figure 2. Vaccine strains are distributed in clusters and named after the earliest vaccine strain. Phylogenetic trees are based on genetic sequence changes, while antigenic clusters reflect more phenotypic differences in antigenic epitope changes.

[0153] Figure 5 The focus of the left figure is to visualize the distribution of antigenic clusters and highlight the name of the vaccine strain in each cluster. These clusters reflect important trends in the changes in viral antigenicity and simplify the complex information in the phylogenetic tree. The right figure shows the phylogenetic relationship of all virus strains, reflecting the genetic continuity and evolution process between virus strains. In actual situations, antigenic changes and genetic changes are not necessarily completely linear, and there may be small genetic changes but significant antigenic changes.

[0154] Figure 6A heat map of intra-cluster and inter-cluster similarities of an antigen cluster is shown. From the similarity heat map, it can be seen that the virus strains within the cluster are verified to have higher antigenic similarity than the virus strains between clusters (the darker the color, the higher the similarity value), indicating the effectiveness of the cluster.

[0155] Step S54: using Markov clustering to perform network clustering on the antigen correlation network to identify virus clusters with similar antigens in the antigen correlation network, and obtain multiple antigen clusters;

[0156] The Markov Clustering Algorithm (MCL) was used for network clustering to identify virus clusters with similar antigenicity in the antigen correlation network. During the clustering process, the edge weight was set as the logarithm of the probability ratio of antigen change to ensure the rationality of virus clustering. Clustering parameters include average cluster size, modularity, etc. By adjusting these parameters, the clustering effect can be optimized and reasonable antigen cluster division results can be obtained.

[0157] Step S55: performing antigenic evolution analysis of the H3N2 influenza virus based on multiple antigenic clusters to obtain corresponding antigenic evolution analysis results.

[0158] Figure 7 Figure 2 shows the average cluster size and modularity of a network cluster as the expansion parameter changes. Figure 7 As shown, blue represents the average cluster size, corresponding to the left y-axis. Green represents the modularity, corresponding to the right y-axis. Figure 7 , the parameter corresponding to the first point of the first plateau, the higher the modularity, the better the clustering effect. Therefore, the expansion parameter is selected as 1.4.

[0159] Through cluster analysis, multiple antigenic clusters were obtained, and each cluster contains a group of viruses with similar antigenicity. The division of these antigenic clusters provides data support for analyzing the antigenic evolution and transmission path of the H3N2 virus. On the basis of the antigenic clusters, by tracking the spatiotemporal distribution of each antigenic cluster, observing the replacement process of old and new antigenic clusters, analyzing the emergence and spread trends of different antigenic clusters around the world or on each continent, and comparing the similarities and differences in the evolution of hemagglutinin (HA) and neuraminidase (NA). The schematic diagram of the spatiotemporal epidemic map of H3N2 virus based on neuraminidase and the comparison based on hemagglutinin is shown in the figure below. Figure 8 As shown. A comparative analysis of the hemagglutinin and neuraminidase of the virus in the antigenic cluster was conducted, and the co-evolution phenomenon of the two in the process of antigenic evolution was found. Compared with the rapid mutation of hemagglutinin, the antigenic changes of neuraminidase are more stable. Thus, through this multi-level antigenic evolution analysis, the present invention not only reveals the antigenic evolution pattern of H3N2 virus, but also provides an important scientific basis for vaccine selection and antigen prediction.

[0160] It should be noted that the present invention uses Cytoscape software to construct and visualize the antigen correlation network. In addition, alternatives can be used, such as network construction tools (Gephi or NetworkX) to replace Cytoscape for visualization and topological analysis of antigen networks. Among them, Gephi is more suitable for interactive analysis, and NetworkX is suitable for scripted processing. Other network analysis algorithms can also be used, such as Node2Vec or GraphSAGE algorithms based on graph embedding to further explore the hidden antigen relationships between virus strains. The formation and propagation trends of antigen clusters can also be dynamically displayed through dynamic network models, such as time series network models. It is understandable that the present invention is not limited to this.

[0161] In an embodiment of the present invention, a method for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus based on neuraminidase is provided. By fully integrating multiple key features of neuraminidase (such as epitope characteristics, physicochemical properties, glycosylation sites and catalytic sites), a high-precision antigenic relationship prediction model based on machine learning is developed based on an integrated learning method of multiple machine learning models to achieve high-precision prediction of antigenic relationships between virus strains. Based on the prediction results, an antigen correlation network covering global H3N2 NA data is further constructed. Through dynamic network and cluster analysis, it is helpful to quickly identify possible new epidemic strains when viral antigens mutate, reveal the distribution and evolution of antigen clusters, and provide new scientific basis for influenza monitoring and vaccine selection. Thus, the technical solution provided by the present invention is adopted to realize the dynamic evolution analysis of antigen clusters on the basis of H3N2 antigenicity prediction, and the formation of new antigen clusters can be tracked and their diffusion trend can be predicted. In addition, the antigen similarity prediction model and network analysis method can be extended to whole genome data or other viral protein antigenicity analysis, which is more adaptable to the complexity and diversity of viral antigen changes.

[0162] Reference Fig. 9 , shows a structural block diagram of a device for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus provided by an embodiment of the present invention, which may specifically include:

[0163] A data acquisition unit 901 is used to acquire multiple neuraminidase virus pairs and multiple effective antigen features of H3N2 influenza virus;

[0164] A feature quantification unit 902 is used to perform feature quantification on the multiple neuraminidase virus pairs based on the multiple effective antigen features to construct an antigen relationship prediction data set;

[0165] A model training unit 903 is used to perform model training on a pre-built two-layer prediction model according to the antigen relationship prediction data set to obtain an antigen similarity prediction model;

[0166] The antigen relationship prediction unit 904 is used to input the neuraminidase sequence data to be tested into the antigen similarity prediction model to perform antigen relationship prediction;

[0167] The antigen evolution analysis unit 905 is used to construct an antigen correlation network based on the antigen relationship prediction result, and perform network clustering on the antigen correlation network to obtain the antigen evolution analysis result of the H3N2 influenza virus.

[0168] In an optional embodiment, the data acquisition unit 901 includes:

[0169] Influenza virus data collection unit, used to collect neuraminidase sequence data and neuraminidase inhibition test data of H3N2 influenza virus;

[0170] An effective antigen feature extraction unit is used to perform sequence screening and double optimization on the neuraminidase sequence data to obtain optimized sequence data, and perform feature selection based on the optimized sequence data to extract multiple effective antigen features;

[0171] The neuraminidase virus pair generating unit is used to first deduplicate the neuraminidase inhibition test data, and then perform antigen distance quantification processing to obtain multiple neuraminidase virus pairs.

[0172] In an optional embodiment, the effective antigen feature extraction unit includes:

[0173] A uniformly distributed random sampling unit, used for performing uniformly distributed random sampling on the neuraminidase sequence data to obtain sampled neuraminidase sequence data;

[0174] A multiple sequence alignment unit, used to perform multiple sequence alignment on the sampled neuraminidase sequence data, perform sequence alignment based on the multiple sequence alignment results, and screen out low-quality neuraminidase sequences whose vacancies exceed a preset vacancy threshold during the sequence alignment process;

[0175] A target neuraminidase sequence screening unit, used to select at least one target neuraminidase sequence with a vacancy from all retained neuraminidase sequences;

[0176] An amino acid empty site determination unit, for determining the amino acid empty site of each target neuraminidase sequence, and recording the sequence context amino acids upstream and downstream of the amino acid empty site;

[0177] A target complete neuraminidase sequence screening unit, used to screen out at least one target complete neuraminidase sequence having a similarity match degree with the sequence context amino acid higher than a preset match degree threshold from all the retained complete neuraminidase sequences;

[0178] A site dynamic filling unit, used to predict the gap filling value of the amino acid gap site by Bayesian interpolation based on the at least one target complete neuraminidase sequence, and dynamically fill the target neuraminidase sequence based on the gap filling value to obtain a filled neuraminidase sequence;

[0179] The sequence data integration unit is used to integrate all the filled neuraminidase sequences and all the complete neuraminidase sequences, and add the H3N2 recommended vaccine strains of previous years to obtain optimized sequence data.

[0180] In an optional embodiment, the effective antigen feature extraction unit includes:

[0181] A sequence structure analysis unit, used to perform structure analysis on each neuraminidase sequence in the optimized sequence data using a plurality of different prediction tools, and record all residues predicted to be antigenic epitopes;

[0182] A potential structural epitope screening unit, used to calculate the number of times each of the residues is predicted to be an antigenic epitope, and to take the residues whose antigenic epitope prediction times are greater than or equal to a preset times threshold as potential structural epitopes;

[0183] An antigen epitope feature determination unit is used to perform cluster analysis based on all the potential structural epitopes, determine the number of clusters according to the Euclidean distance relationship and silhouette coefficient between each of the potential structural epitopes, and obtain five antigen epitope features;

[0184] A sequence residue screening unit, used for screening hydrophobic residues, charge distribution residues, charge polarity residues, molecular volume residues and solvent accessible surface area residues from each neuraminidase sequence of the optimized sequence data;

[0185] a physicochemical property feature determination unit, for performing feature selection through a random forest model based on all the hydrophobic residues, the charge distribution residues, the charge polarity residues, the molecular volume residues and the solvent accessible surface area residues, and screening out five physicochemical property features representing hydrophobicity, charge, polarity, volume and accessible surface area, respectively, according to feature importance scores;

[0186] An N-glycosylation site feature determination unit is used to predict the N-glycosylation site of each neuraminidase sequence in the optimized sequence data to obtain the N-glycosylation site feature;

[0187] The enzyme catalytic site feature determination unit is used to determine the enzyme catalytic site features of H3N2 influenza virus neuraminidase based on data collection and analysis.

[0188] In an optional embodiment, the feature quantization unit 902 includes:

[0189] an antigen epitope feature quantification unit, for comparing, for each antigen epitope feature, whether the amino acids of each site of the neuraminidase virus pair in the corresponding epitope region are different, and taking the number of differences as the antigen epitope feature value of the neuraminidase virus pair in the antigen epitope feature;

[0190] A physicochemical property characteristic quantification unit, for calculating the difference value of the neuraminidase virus pair at each site for each physicochemical property characteristic, and selecting the three largest difference values ​​to solve the average value as the physicochemical property characteristic value of the neuraminidase virus pair in the physicochemical property characteristic;

[0191] An N-glycosylation site characteristic quantification unit, used to calculate the number of differences in the N-glycosylation sites of the neuraminidase virus pair as the N-glycosylation characteristic value of the neuraminidase virus pair at the N-glycosylation site characteristic;

[0192] An enzyme catalytic site characteristic quantification unit, used to calculate the Euclidean distance value from each differential site in the neuraminidase virus pair to the enzyme catalytic site, and select the three largest Euclidean distance values ​​to solve the average value as the enzyme catalytic site characteristic value of the neuraminidase virus pair at the enzyme catalytic site characteristic;

[0193] The antigenic relationship prediction data set construction unit is used to integrate the antigenic epitope characteristic values, physicochemical property characteristic values, N-glycosylation characteristic values ​​and enzyme catalytic site characteristic values ​​of all the neuraminidase virus pairs to construct an antigenic relationship prediction data set.

[0194] In an optional embodiment, the two-layer prediction model includes a first-layer basic prediction model and a second-layer comprehensive prediction model; the first-layer basic prediction model includes a plurality of different prediction sub-models; the model training unit 903 includes:

[0195] A data set division unit, used for dividing the antigen relationship prediction data set into an antigen relationship prediction training set and an antigen relationship prediction test set according to a preset division ratio;

[0196] A prediction sub-model training unit, used for performing model training on each of the prediction sub-models respectively through the antigen relationship prediction training set;

[0197] A comprehensive prediction model training unit, used for inputting the prediction results of each prediction sub-model into the second-layer comprehensive prediction model to perform model training based on comprehensive prediction;

[0198] A model parameter optimization unit, used to perform layered training on the prediction results of each prediction sub-model by cross-validation during the model training process, and optimize the model parameters of the two-layer comprehensive prediction model and each prediction sub-model by grid parameter adjustment and random parameter adjustment;

[0199] The model evaluation unit is used to use the antigen relationship prediction test set to perform model evaluation on the trained two-layer comprehensive prediction model and each trained prediction sub-model, and output the final antigen similarity prediction model.

[0200] In an optional embodiment, the antigen relationship prediction result is that the H3N2 neuraminidase virus pair in the neuraminidase inhibition test data is antigenically similar, or antigenically dissimilar; the antigen evolution analysis unit 905 includes:

[0201] A network visualization unit, used for performing network visualization on the antigen relationship prediction result;

[0202] An antigen correlation network construction unit is used to take each virus strain in the neuraminidase sequence data to be tested as a node, and take the antigen similarity relationship between virus strains as node edges, and construct an antigen correlation network of H3N2 influenza virus based on each of the nodes and each of the node edges;

[0203] A phylogenetic tree construction unit, used to construct a phylogenetic tree of H3N2 influenza virus based on the antigen correlation network, wherein the phylogenetic tree is used to reflect the genetic continuity and evolution process between virus strains;

[0204] A network clustering unit, used for performing network clustering on the antigen correlation network by using Markov clustering to identify virus clusters with similar antigens in the antigen correlation network to obtain multiple antigen clusters;

[0205] The antigen evolution analysis subunit is used to perform antigen evolution analysis of the H3N2 influenza virus based on the multiple antigen clusters to obtain corresponding antigen evolution analysis results.

[0206] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the aforementioned method embodiment.

[0207] An embodiment of the present invention further provides an electronic device, the device comprising a processor and a memory:

[0208] The memory is used to store the program code and transmit the program code to the processor;

[0209] The processor is used to execute the method for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus according to any embodiment of the present invention according to the instructions in the program code.

[0210] An embodiment of the present invention further provides a computer-readable storage medium, which is used to store program code, and the program code is used to execute the method for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus according to any embodiment of the present invention.

[0211] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0212] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0213] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0214] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0215] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0216] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus, characterized in that: include: Obtain multiple neuraminidase virus pairs and multiple effective antigenic features of H3N2 influenza virus; quantifying the features of the plurality of neuraminidase virus pairs based on the plurality of effective antigen features to construct an antigen relationship prediction data set; Performing model training on a pre-constructed two-layer prediction model according to the antigen relationship prediction data set to obtain an antigen similarity prediction model; Inputting the neuraminidase sequence data to be tested into the antigen similarity prediction model to perform antigen relationship prediction; An antigen correlation network was constructed based on the antigen relationship prediction results, and network clustering was performed on the antigen correlation network to obtain the antigen evolution analysis results of the H3N2 influenza virus.

2. The method for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus according to claim 1, characterized in that: The method of obtaining multiple neuraminidase virus pairs and multiple effective antigenic features of H3N2 influenza virus includes: Collect neuraminidase sequence data and neuraminidase inhibition test data of H3N2 influenza virus; Performing double optimization of sequence screening on the neuraminidase sequence data to obtain optimized sequence data, and performing feature selection based on the optimized sequence data to extract multiple effective antigen features; The neuraminidase inhibition test data is first deduplicated, and then antigen distance quantification is performed to obtain multiple neuraminidase virus pairs.

3. The method for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus according to claim 2, characterized in that: The step of performing sequence screening and double optimization on the neuraminidase sequence data to obtain optimized sequence data comprises: Performing uniformly distributed random sampling on the neuraminidase sequence data to obtain sampled neuraminidase sequence data; Performing a multiple sequence alignment on the sampled neuraminidase sequence data, performing sequence alignment based on the multiple sequence alignment results, and screening out low-quality neuraminidase sequences whose vacancies exceed a preset vacancy threshold during the sequence alignment process; Select at least one target neuraminidase sequence with a gap from all retained neuraminidase sequences; For each of the target neuraminidase sequences, determine the amino acid vacancy site of the target neuraminidase sequence, and record the sequence context amino acids upstream and downstream of the amino acid vacancy site; Screening out, from all the retained complete neuraminidase sequences, at least one target complete neuraminidase sequence having an amino acid similarity match degree with the sequence context higher than a preset match degree threshold; Based on the at least one target complete neuraminidase sequence, predicting the gap filling value of the amino acid gap site by Bayesian interpolation, and dynamically filling the target neuraminidase sequence based on the gap filling value to obtain a filled neuraminidase sequence; All the filled neuraminidase sequences and all the complete neuraminidase sequences were integrated, and the H3N2 recommended vaccine strains of previous years were added to obtain optimized sequence data.

4. The method for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus according to claim 2, characterized in that: The feature selection based on the optimized sequence data to extract multiple effective antigen features includes: Using a variety of different prediction tools to perform structural analysis on each neuraminidase sequence in the optimized sequence data, and recording all residues predicted to be antigenic epitopes; Calculating the number of times each of the residues is predicted to be an antigenic epitope, and taking the residues whose antigenic epitope prediction times are greater than or equal to a preset times threshold as potential structural epitopes; Performing cluster analysis based on all the potential structural epitopes, determining the number of clusters according to the Euclidean distance relationship and silhouette coefficient between each of the potential structural epitopes, and obtaining five antigen epitope features; Screening hydrophobic residues, charge distribution residues, charge polarity residues, molecular volume residues and solvent accessible surface area residues from each neuraminidase sequence in the optimized sequence data; According to all the hydrophobic residues, the charge distribution residues, the charge polarity residues, the molecular volume residues and the solvent accessible surface area residues, feature selection is performed by a random forest model, and five physicochemical property features representing hydrophobicity, charge, polarity, volume and accessible surface area are screened out according to feature importance scores; Predicting the N-glycosylation site of each neuraminidase sequence in the optimized sequence data to obtain N-glycosylation site characteristics; Based on data collection and analysis, the enzyme catalytic site characteristics of H3N2 influenza virus neuraminidase were determined.

5. The method for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus according to claim 4, characterized in that: The step of quantifying the characteristics of the plurality of neuraminidase virus pairs based on the plurality of effective antigen characteristics and constructing an antigen relationship prediction data set comprises: For each antigenic epitope feature, comparing whether the amino acids of each site of the neuraminidase virus pair in the corresponding epitope region are different, and taking the number of differences as the antigenic epitope feature value of the neuraminidase virus pair in the antigenic epitope feature; For each physicochemical property characteristic, the difference value of the neuraminidase virus pair at each site is calculated, and the three largest difference values ​​are selected to calculate the average value as the physicochemical property characteristic value of the neuraminidase virus pair in the physicochemical property characteristic; Calculating the number of differences in the N-glycosylation sites of the neuraminidase virus pair as the N-glycosylation characteristic value of the neuraminidase virus pair at the N-glycosylation site; Calculate the Euclidean distance value from each differential site to the enzyme catalytic site in the neuraminidase virus pair, and select the three largest Euclidean distance values ​​to calculate the average value as the enzyme catalytic site characteristic value of the neuraminidase virus pair at the enzyme catalytic site characteristic; The antigenic epitope characteristic values, physicochemical property characteristic values, N-glycosylation characteristic values ​​and enzyme catalytic site characteristic values ​​of all the neuraminidase virus pairs were integrated to construct an antigenic relationship prediction data set.

6. The method for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus according to any one of claims 1 to 5, characterized in that: The two-layer prediction model includes a first-layer basic prediction model and a second-layer comprehensive prediction model; the first-layer basic prediction model includes a plurality of different prediction sub-models; the pre-constructed two-layer prediction model is trained according to the antigen relationship prediction data set to obtain an antigen similarity prediction model, including: Dividing the antigen relationship prediction data set into an antigen relationship prediction training set and an antigen relationship prediction test set according to a preset division ratio; Performing model training on each of the prediction sub-models respectively using the antigen relationship prediction training set; Inputting the prediction results of each prediction sub-model into the second-layer comprehensive prediction model to perform model training based on comprehensive prediction; During the model training process, cross-validation is used to perform hierarchical training on the prediction results of each prediction sub-model, and the model parameters of the two-layer comprehensive prediction model and each prediction sub-model are optimized respectively through grid parameter adjustment and random parameter adjustment; The antigen relationship prediction test set is used to perform model evaluation on the trained two-layer comprehensive prediction model and each trained prediction sub-model, and the final antigen similarity prediction model is output.

7. The method for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus according to claim 6, characterized in that: The antigenic relationship prediction result is that the H3N2 neuraminidase virus pair in the neuraminidase inhibition test data is antigenically similar, or antigenically dissimilar; the antigenic correlation network is constructed based on the antigenic relationship prediction result, and the antigenic correlation network is clustered to obtain the antigenic evolution analysis result of the H3N2 influenza virus, including: Performing network visualization on the antigen relationship prediction results; Taking each virus strain in the neuraminidase sequence data to be tested as a node, and taking the antigenic similarity relationship between virus strains as node edges, and constructing an antigen correlation network of H3N2 influenza virus based on each of the nodes and each of the node edges; Constructing a phylogenetic tree of H3N2 influenza virus based on the antigen correlation network, wherein the phylogenetic tree is used to reflect the genetic continuity and evolution process between virus strains; Performing network clustering on the antigen correlation network using Markov clustering to identify virus clusters with similar antigens in the antigen correlation network to obtain multiple antigen clusters; Based on the multiple antigen clusters, antigen evolution analysis of H3N2 influenza virus is performed to obtain corresponding antigen evolution analysis results.

8. A device for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus, characterized in that: include: A data acquisition unit, used to acquire multiple neuraminidase virus pairs and multiple effective antigen features of H3N2 influenza virus; A feature quantification unit, configured to perform feature quantification on the plurality of neuraminidase virus pairs based on the plurality of effective antigen features, and construct an antigen relationship prediction data set; A model training unit, used to perform model training on a pre-built two-layer prediction model according to the antigen relationship prediction data set to obtain an antigen similarity prediction model; An antigen relationship prediction unit, used for inputting the sequence data of the neuraminidase to be tested into the antigen similarity prediction model to perform antigen relationship prediction; The antigen evolution analysis unit is used to construct an antigen correlation network based on the antigen relationship prediction result, and perform network clustering on the antigen correlation network to obtain the antigen evolution analysis result of the H3N2 influenza virus.

9. An electronic device, characterized in that: The device comprises a processor and a memory: The memory is used to store program code and transmit the program code to the processor; The processor is used to execute the method for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus according to any one of claims 1 to 7 according to the instructions in the program code.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store program codes, and the program codes are used to execute the method for analyzing the dynamic evolution of antigenicity of H3N2 influenza virus according to any one of claims 1 to 7.

Images