The journey so far with SARS-CoV-2 variants: Pathogenesis, immunity and treatments

Daniel Danladi Gaiya; Jonathan Danladi Gaiya; Richard Auta; Aliyu Muhammad; Bege Jonathan; Stella Kuyet Udu; Ekpa Emmanuel; Amina Shehu Bature; Daniel Danladi Gaiya; Jonathan Danladi Gaiya; Richard Auta; Aliyu Muhammad; Bege Jonathan; Stella Kuyet Udu; Ekpa Emmanuel; Amina Shehu Bature

doi:10.3934/Allergy.2023016

AIMS Allergy and Immunology

2023, Volume 7, Issue 4: 222-250. doi: 10.3934/Allergy.2023016

Previous Article Next Article

Review Special Issues

The journey so far with SARS-CoV-2 variants: Pathogenesis, immunity and treatments

1.
Biology Unit, Air Force Institute of Technology, Nigerian Air Force Base, Kawo, Kaduna, Nigeria
2.
Polymer Technology Department, Nigeria Institute of Leather & Science Technology, Zaria, Nigeria
3.
Department of Chemical Engineering, Ahmadu Bello University, Samaru Zaria, Nigeria
4.
Department of Biochemistry, Faculty of Science, Kaduna State University, Kaduna, Nigeria
5.
Department of Biochemistry, Faculty of Life Sciences, Ahmadu Bello University, Samaru Zaria, Nigeria
6.
Department of Biochemistry, Faculty of Natural and Applied Sciences, Plateau State University, Bokkos, Jos, Nigeria

Received: 29 April 2023 Revised: 22 August 2023 Accepted: 24 August 2023 Published: 15 September 2023

The recruitment of therapeutics and most importantly COVID-19 vaccines has seen a measurable reduction in transmission, re-infection, severity, hospitalization and mortality associated with the novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). The development and approval of some vaccines and therapeutics undoubtedly signaled renewed hope for public health personnel, the government, the World Health Organization (WHO) and the entire world population. At present, most countries have progressed beyond administering first and second doses to administering COVID-19 vaccine updated boosters to prevent transmission and provide protection. Notably, a bivalent COVID-19 vaccine from Pfizer–BioNTech and Moderna, also called an “updated” COVID-19 vaccine booster dose, is a formulation that houses the original virus strain and omicron BA.1, which provides broad immunity against COVID-19 including the omicron variant (BA.1) and the Paxlovid drug (Nirmatrelvir-ritonavir) authorized for use by the Food and Drug Administration (FDA) and the European Medicines Agency. This current review outlines the variant of concern (VOC), viral cell entry and pathogenesis, host immunity and viral immune evasion. In addition, we discuss the therapeutic and vaccine treatment approach, WHO and FDA authorization, vaccine storage and vaccine efficacy. In conclusion, bearing in mind the trend of continued mutations as observed on the spike (S) glycoprotein and receptor binding domain (RBD) of SARS-CoV-2, which lead to more immune-evasive strains such as BQ.1, BQ.1.1, BF.7, XBB and XBB.1, researcher and clinician attention should be tailored toward the design and development of variant-specific vaccines for future interventions.

Keywords:

Citation: Daniel Danladi Gaiya, Jonathan Danladi Gaiya, Richard Auta, Aliyu Muhammad, Bege Jonathan, Stella Kuyet Udu, Ekpa Emmanuel, Amina Shehu Bature. The journey so far with SARS-CoV-2 variants: Pathogenesis, immunity and treatments[J]. AIMS Allergy and Immunology, 2023, 7(4): 222-250. doi: 10.3934/Allergy.2023016

Related Papers:

[1]	Yue Li, Wusheng Xu, Wei Li, Ang Li, Zengjin Liu . Research on hybrid intrusion detection method based on the ADASYN and ID3 algorithms. Mathematical Biosciences and Engineering, 2022, 19(2): 2030-2042. doi: 10.3934/mbe.2022095
[2]	Hao Zhang, Lina Ge, Guifen Zhang, Jingwei Fan, Denghui Li, Chenyang Xu . A two-stage intrusion detection method based on light gradient boosting machine and autoencoder. Mathematical Biosciences and Engineering, 2023, 20(4): 6966-6992. doi: 10.3934/mbe.2023301
[3]	Naila Naz, Muazzam A Khan, Suliman A. Alsuhibany, Muhammad Diyan, Zhiyuan Tan, Muhammad Almas Khan, Jawad Ahmad . Ensemble learning-based IDS for sensors telemetry data in IoT networks. Mathematical Biosciences and Engineering, 2022, 19(10): 10550-10580. doi: 10.3934/mbe.2022493
[4]	Zhongwei Li, Wenqi Jiang, Xiaosheng Liu, Kai Tan, Xianji Jin, Ming Yang . GAN model using field fuzz mutation for in-vehicle CAN bus intrusion detection. Mathematical Biosciences and Engineering, 2022, 19(7): 6996-7018. doi: 10.3934/mbe.2022330
[5]	Muhammad Hassan Jamal, Muazzam A Khan, Safi Ullah, Mohammed S. Alshehri, Sultan Almakdi, Umer Rashid, Abdulwahab Alazeb, Jawad Ahmad . Multi-step attack detection in industrial networks using a hybrid deep learning architecture. Mathematical Biosciences and Engineering, 2023, 20(8): 13824-13848. doi: 10.3934/mbe.2023615
[6]	Noor Wali Khan, Mohammed S. Alshehri, Muazzam A Khan, Sultan Almakdi, Naghmeh Moradpoor, Abdulwahab Alazeb, Safi Ullah, Naila Naz, Jawad Ahmad . A hybrid deep learning-based intrusion detection system for IoT networks. Mathematical Biosciences and Engineering, 2023, 20(8): 13491-13520. doi: 10.3934/mbe.2023602
[7]	Kunpeng Li, Zepeng Wang, Yu Zhou, Sihai Li . Lung adenocarcinoma identification based on hybrid feature selections and attentional convolutional neural networks. Mathematical Biosciences and Engineering, 2024, 21(2): 2991-3015. doi: 10.3934/mbe.2024133
[8]	Jing Zhang, Baoqun Yin, Yu Zhong, Qiang Wei, Jia Zhao, Hazrat Bilal . A two-stage grasp detection method for sequential robotic grasping in stacking scenarios. Mathematical Biosciences and Engineering, 2024, 21(2): 3448-3472. doi: 10.3934/mbe.2024152
[9]	Ziyu Jin, Ning Li . Diagnosis of each main coronary artery stenosis based on whale optimization algorithm and stacking model. Mathematical Biosciences and Engineering, 2022, 19(5): 4568-4591. doi: 10.3934/mbe.2022211
[10]	Natalya Shakhovska, Vitaliy Yakovyna, Valentyna Chopyak . A new hybrid ensemble machine-learning model for severity risk assessment and post-COVID prediction system. Mathematical Biosciences and Engineering, 2022, 19(6): 6102-6123. doi: 10.3934/mbe.2022285

Abstract

1. Introduction

With the development of internet technology, there is a tremendous increase in the dimensions of information that are generated, exchanged and processed. Almost in all the fields there is a difficulty in handling large amount of high dimensional data. These data become the target for illegal activities which impose a severe threat to the network security. The traditional security techniques such as antivirus, firewall, data encryption and user identification which acts as the first line of defense that alone is not sufficient to provide the better security to the network. The second line of security is highly recommended that can be provide by an Intrusion Detection System (IDS) ^[1]. Using these two lines of security enhances the overall network security. An Intrusion Detection System (IDS) identifies the intruders who are harmful to an organization. The main goal of an IDS is to monitor the network or system for abnormal pattern or traffic and prevent it from unauthorized access ^[2]. However, the key problem is identifying the unknown malicious traffic. The Intrusion Detection Systems is broadly classified into two types based on the source of data. First, Network Intrusion Detection System (NIDS) placed near the network points checks the network traffic from the routers and gateways for intrusion. It can detect attacks in real time. The main limitation of NIDS is that it can monitor the traffic passing through the specific network nodes. Second, the Host Intrusion Detection System (HIDS) scans the individual host for suspicious activities like an unwanted configuration change, deletion or modification of system files, or an unwanted sequence of system calls. If any of these activity occurs, it sends an alert to the administrator ^[3]. The main limitation in HIDS is that it cannot analyze the network behaviors. Next the Intrusion Detection System (IDS) uses three different detection mechanisms to detect the attacks and each detection method is further classified accordingly. Figure 1 shows the different detection mechanisms to detect the intrusions.

Figure 1. Types of detection mechanism.

DownLoad: Full-Size Img PowerPoint

1.1. Signature-based/misuse detection mechanism

Signature-based IDS uses the set of rules or predefined signatures for detecting the known attacks ^[3]. Misuse detection techniques are based on knowledge or based on Machine Learning (ML) methods. In a knowledge-based strategy, the network stream of traffic or host audit data is analyzed and compared with the predefined set of rules. There are three ways of applying the knowledge-based approach: signature matching, rule-based expert system, and state transition analysis ^[3]. The signature matching compares the incoming network traffic with the predefined attack signatures for intrusion. A rule-based expert system finds out the intrusion by comparing the traffic with the predetermined rules. Finally, state transition analysis maintains the state transition model for each known suspicious activity. Machine Learning (ML) based IDS offers learning based on model to detect the normal behavior from attack behavior. The ML model generates the representation for the known models. It uses supervised machine learning techniques such as SVM, Decision Tree to detect the known attacks more efficiently.

The significant drawback of Signature-based Detection is that it needs to update the signatures regularly for new attacks for which signatures are not there in the database. As a result, it generates more false alarms. It has to maintain a large signature database.

1.2. Anomaly-based detection mechanism

Anomaly-based detection uses a hypothesis to detect novel, unknown intrusion if any deviation or behaviour change occurs ^[3]. Anomaly-based detection comprises statistical techniques, Finite State Machine (FSM) and machine learning techniques. FSM generates the behaviour model that contains states, actions and transitions. Semi-Supervised and Unsupervised ML techniques such as clustering algorithms, one class SVM are mainly used for anomaly detection. The anomaly-based detection can detect both known and unknown attacks. The prime limitation is that anomaly-based detection suffers from high false positives ^[3].

1.3. Hybrid detection mechanism

Hybrid detection mechanism combines signature and anomaly-based detection to detect the intrusion ^[3].

1.4. Limitations of current IDS

Current IDS can detect more accurately and precisely the known attacks, which leaves the system more vulnerable to novel malicious attacks where predefined signatures are not available ^[4].

High False positives-The current IDS suffers from high false positives. False positives are the incorrect classification of a normal event as malicious events. The main aim of an IDS is to minimize the false positives as far as possible ^[4].

High False-negatives-The current IDS suffers from high false negatives. False negatives are the incorrect classification of malicious events as normal events. The main aim of an IDS is to reduce the false negatives as minimum as possible ^[4].

Data Overload-Millions of data are generated every day depending on the company's size and the IDS tools used, which leads to a data overload ^[4].

1.5. DataStream mining

DataStream mining is the continuous well-ordered sequence of data that arrives in a timely fashion ^[5]. Data streams are a constant flow of unlimited data with high speed, and data changes with time compared to the traditional databases ^[6]. The primary requirement of the stream data is to inspect each instance only once. Therefore, it should use a limited amount of memory and should give the outcome in a less amount of time. Data streams are of two types, namely streams that are static and streams that are evolving. The bulk arrival of data that won't change with time is called static stream. The data that arrives continuously and changes with time is called evolving data streams ^[6].

The paper is organized as follows: Section 2 presents the earlier work on various machine learning and feature selection techniques. Section 3 discusses the methodology used for preprocessing, feature selection and classification. Section 4 discusses the experimental setup, dataset details and the performance evaluation metrics used. Section 5 presents the result outcomes and its performance comparison with the state of art methods and finally Section 6 summarizes the research work undergone and future scope.

2. Related works

There has been quite a lot of research on intrusion detection using machine learning and deep learning techniques and the hybrid approaches. The related work (Table 1) presents the latest techniques used and its relevant advantages and disadvantages.

Table 1. Related works.

Authors	Algorithms Used	Computational Methods	Pros	Cons
X. Li et al. (2021) ^[7]	CMPSO, ACO, KH, IKH, LNNLS-KH	The NSL_KDD dataset is taken for intrusion detection. The proposed LNNLS-KH is compared with CMPSO, ACO, KH, and IKH algorithms and best features are selected. Then KNN technique is applied for further classification.	Good convergence speed. Lower false positive rate. LNNLS-KH gives an accuracy of 96.12%	Attack with fewer samples, an adversarial learning method can be used to create similar attacks.
X. K. Zhou et al. (2021) ^[8]	AdaBoost, LSTM, CNN LSTM, VLSTM	The authors proposed a variational long short-term memory technique that detects intrusion anomalies efficiently based on feature reconstruction.	The loss function helps to reconstruct the hidden variable into meaningful form. Proposed VLSTM gives the accuracy of 89.5%.	Imbalanced data is still a challenge in anomaly detection.
T. H. Hai et al. (2020) ^[9]	Novel architecture of storage tools and distributed log processing	The Novel storage with HBase or Apache Spark enhances NIDS data processing.	Processing time is reduced.	Takes more query time.
S. N. Mighan et al. (2020) ^[10]	SVM, SAE-SVM	Hybrid scheme that uses deep learning and machine learning method together (SAE-SVM) can detect intrusion attacks more precisely.	SAE-SVM shows higher accuracy of 95.98%.	Computational time taken is more.
T. Vaiyapuri et al. (2020) ^[11]	Stacked Autoencoder, Sparse Autoencoder, Denoising Autoencoder, Contractive Autoencoder, Convolution Autoencoder	All the methods mentioned are compared with contractive autoencoder. The contractive autoencoder gives 87.98% intrusion detection accuracy.	SAE 85.23%, SAAE 86.02% DAE 86.92% ContAE 87.98% CAE 81.07%	Reduced reconstruction ability further needs to be improved
C. F. Tang et al. (2020) ^[12]	Stacked Autoencoder Deep Neural Network (SAE-DNN), SAAE-DNN	SAAE selects the needed features from the intrusion dataset and initializes weight to DNN thus improves the intrusion detection accuracy.	SAE-DNN gives 82.23% accuracy. The proposed SAAE-DNN shows higher accuracy of 87.74% when compared to SAE-DNN.	The accuracy achieved can be improved further.
A. D. Jadhav et al. (2019) ^[13]	SVM, KNN, Decision Tree, Naïve Bayes classifier machine learning techniques to detect the attacks. Proposed a distributed and parallel approach.	The proposed distributed and parallel approach enhances the efficiency of detecting the intrusions faster.	Faster detection of intrusion.	Not applied in real time environment.
A. Muallem et al. (2017) ^[14]	Hoeffding Tree Restricted, Hoeffding Trees AUE2 with Buffer AUE2 without Buffer Hoeffding Adaptive Trees with DDM and ADWIN	The survey talks about the flexibility of the methods when used in different domains of streaming data. Combinations of technique solves the intrusion anomaly detection problem.	All the techniques surveyed gives good accuracy. Hoeffding Tree 93% Restricted Hoeffding Trees 92.15% AUE2 with Buffer 94.07% AUE2 without Buffer 94.06% Hoeffding Adaptive Trees with DDM and ADWIN 92%.	Improvement in accuracy is needed.
G. Kim et al. (2014) ^[15]	Hybrid approach combines C 4.5 algorithm and one class SVM approach	C 4.5 is used to build the misuse detection model and decompose the training data into smaller subsets. Multiple one class SVM models are built to enhance intrusion detection accuracy	Good detection accuracy for both known and unknown attacks. Low false positive.	Processing time is more for the proposed technique.
H. K. Sok et al. (2013) ^[16]	ADT algorithm	The alternating decision tree algorithm is used for knowledge discovery and effective selection of features.	Classification process simplifies.	Evaluation speed needs to be improved.
S. J. Horng et al. (2011) ^[17]	BIRCH hierarchical clustering technique, SVM	It combines BIRCH hierarchical clustering technique and SVM which gives good detection accuracy.	The training time is reduced and gives a good detection accuracy of 95.72%. It mainly detects DOS or Probe attacks.	It cannot detect U2L and R2L attacks.
Tavallaee et al. (2009) ^[18]	KDD CUP 99 dataset	Analysis on KDD CUP 99 dataset is made.	Good for signature-based detection.	Poor detection when used for anomaly detection. Contains many duplicate values leads to performance degradation.

| Show Table

DownLoad: CSV

3. Methodology

The proposed work uses the NSL_KDD benchmark dataset for intrusion analysis. The first phase focuses on feature selection using Darwinian Particle Swarm Optimization and selects key features that contribute to intrusion. The second phase emphasises applying the proposed Stacked Encoding Hoeffding Tree technique to classify the data based on performance metrics like accuracy, specificity, sensitivity, false-positive rate, false-negative rate and F1 score.

The benchmark NSL_KDD dataset is taken for the analysis. The dataset which we have taken is standalone dataset in order to incorporate stream data, the dataset is streamed using the techniques in Matlab tool. The system object technique simplifies the streaming process in Matlab. The data is now continuous and it possess the characteristics of streaming data. The attacks are detected using the proposed Stacked Autoencoder Hoeffding Tree (SAE-HT) classification approach. The bio-inspired technique called Darwinian Particle Swarm Optimization (DPSO) enhances the performance of the SAE-HT classification technique. The distracting variance is removed from the data using the DPSO feature selection technique that enables the classifier to perform better, especially when dealing with the high dimensional features. Figure 2 shows the flow diagram of SAE-HT classification technique.

Figure 2. Flow diagram of SAE-HT classification technique.

DownLoad: Full-Size Img PowerPoint

3.1. Data Preprocessing

Data preprocessing is a vital step in machine learning. The raw data collected is made ready to be used by machine learning techniques to extract meaningful insights from the data. The NSL_KDD dataset taken from Canadian Institute for Cybersecurity is analyzed. If the downloaded dataset is in gz or Tcpdump format, convert that to CSV file format and load the dataset into the environment. The proposed classification technique supports only the numeric data. Since most machine learning (ML) techniques use mathematical equations that only support the usage of numeric data, the conversion of categorical data into numerical data using data conversion functions should occur. The one-hot encoding technique converts the categorical data to numerical data, thus making it convenient to apply machine learning techniques to the dataset.

3.2. One-hot encoding technique

The one-hot encoding technique is the most effective encoding technology to deal with the conversion of numeric to categorical features ^[19]. It can convert the categorical features to a binary vector. The vector holds Zeroes, and One's as values. The vector holds only one element with the value one and other values corresponding to Zero. An element with value one indicates the occurrence of the possible values against the categorical features. The NSL_KDD dataset contains three categorical features such as protocol_type, service and flag. For example, the protocol_type consists of three attributes: ICMP, TCP, UDP. Using one hot encoding technique, ICMP can be encoded as (1, 0, 0), TCP can be encoded as (0, 1, 0), UDP can be encoded as (0, 0, 1). Similarly, categorical features service and flag are also encoded into one-hot encoding vectors.

3.3. Feature selection

Feature selection mainly reduces the features by removing the insignificant or less significant ones. There are many feature selection techniques. This paper uses the Bio-Inspired feature selection technique called Darwinian Particle Swarm Optimization (DPSO) to select optimal features. The main goal of the DPSO is to find the non-redundant and highly correlated features, thus eliminates the least correlated features ^[20]. DPSO is the extension of the PSO technique with the basic principle of survival of the fittest. The major drawback of PSO and other bio-inspired feature selection techniques is that it gets trapped in the local optimum. No long-term memory effect leads to premature convergence to the local optimum DPSO overcomes the above drawback. DPSO consists of multiple swarms in which each swarm performs like an individual PSO. All the swarms run simultaneously towards the local optimum, and each swarm is compared. The best swarm gets the extended life, and the stagnated or insufficient swarm got deleted.

DPSO is an effective evolutionary algorithm that searches the population of individuals for local optimum. The population represents the "swarm", and individuals represent the "particles". Throughout the evolutionary process, every particle updates its moving direction according to the position. There are two positions, namely local best and global best position ^[21]. The local best position (pbest) is the particle's position among all the particles that are visited so far. The global best position (gbest) is the best fitness achieved among all the visited particles so far. In each iteration, the particle updates its velocity and position, which is given by Eqs (1) and (2) ^[22].

${V}_{j}\left(t+1\right) = {\omega V}_{j}\left(t\right)+{c}_{1}{r}_{1}\left({P}_{j}-{Y}_{j}\left(t\right)\right)+{c}_{2}{r}_{2}\left({P}_{g}-{Y}_{j}\left(t\right)\right)$

(1)

${Y}_{j}\left(t+1\right) = {Y}_{j}\left(t\right)+{V}_{j}(t+1)$

(2)

${Y}_{j} = ({Y}_{j}^{1}, {Y}_{j}^{2}, \dots., {Y}_{j}^{D})$ denotes the particle position at generation j in a D-dimensional search space. V_j (t + 1) is the velocity produced at time t + 1. P_j is the best position of the particles found so far (pbest). It denotes the cognitive component of Eq (1). P_g is the best global position found so far (gbest). It represents the social component of Eq (1). ω signifies the inertia weight, c signifies a constant called local and global weight, r signifies random variable which ranges between (0, 1). The searching process keeps on going until the predefined threshold reaches. DPSO is more efficient than the original PSO and thus prevents premature convergence to a local optimum. Rearrange the velocity function in Eq (1), and it is given by Eq (3) ^[22].

${V}_{j}\left(t+1\right) = \alpha {V}_{j}\left(t\right)+\frac{\alpha }{2}{V}_{j}\left(t-1\right)+\frac{\alpha \left(1-\alpha \right)}{6}{V}_{j}\left(t-2\right)\\ +\frac{\alpha \left(1-\alpha \right)\left(2-\alpha \right)}{24}{V}_{j}\left(t-3\right)+{c}_{1}{r}_{1}\left({P}_{j}-{Y}_{j}\left(t\right)\right)+{c}_{2}{r}_{2}({P}_{g}-{Y}_{j}\left(t\right))$

(3)

The left side of Eq (3) gives a discrete version of the derivative of velocity D𝛼[V𝑡 + 1] with order α = 1. Finally, the Grunwald-Letnikov derivative expresses the discrete-time implementation using Eq (4) ^[22].

${D}^{\alpha }\left[{V}_{t+1}\right] = \frac{1}{{T}^{\alpha }}\sum \limits_{k = 0}^{r}\frac{{\left(-1\right)}^{k}\rm{\Gamma }\left(\alpha +1\right)v(t-kT)}{\rm{\Gamma }(k+1)\rm{\Gamma }(\alpha -k+1)}$

(4)

Here T represents a sample period, and r represents the truncate order. Repeat Eq (3) to update every particle velocity. The different values generated to control the convergence speed of the optimization process.

3.4. Proposed classification algorithm (SAE-HT)

Classification is the supervised machine learning technique that divides the set of data into different classes. As the data generation is continuous, it is impossible to store a massive amount of data. So, the data needs to be analyzed as it comes in. The classification techniques are very much helpful in classifying the streaming data ^[23]. This paper proposes a classification technique for stream data mining called the Stacked Autoencoder Hoeffding Tree approach. Even though stacked autoencoder gives good accuracy on its own but in order to improve it further we applied hoeffding tree approach. The stacked autoencoder is an unsupervised learning technique that maintains three layers: input, output, and hidden layers ^[24]. An encoder takes the input and maps it to the hidden representation. Finally, the decoder reconstructs the input. Equation (5) gives the encoder process ^[24].

${\rm{h}}_{\rm{n}} = \rm{f}({\rm{W}}_{1}{\rm{x}}_{\rm{n}}+{\rm{b}}_{1})$

(56)

where h_n represents the encoder, vector determined from x_n. Encoding function f, weight matrix of the encoder W₁, and bias vector b₁.

Equation 6 represents the decoder process. Where g denotes decoding function, W₂ denotes the decoders weight matrix, and b₂ denotes the bias vector ^[24].

$\widehat{{\rm{X}}_{\rm{n}}} = \rm{g}({\rm{W}}_{2}{\rm{h}}_{\rm{n}}+{\rm{b}}_{2})$

(6)

When the decoder reconstructs the input data, there is a possibility that it results in reconstruction error. Equation (7) minimizes the reconstruction error ^[24].

$\rm{ \mathsf{ ϕ} }\left(\rm{\Theta }\right) = \genfrac{}{}{0pt}{}{\rm{arg}\rm{m}\rm{i}\rm{n}}{\rm{ \mathsf{ θ} }, {\rm{ \mathsf{ θ} }}^{\rm{\text{'}}}}\frac{1}{\rm{n}}{\sum }_{\rm{i} = 1}^{\rm{n}}\rm{L}({\rm{X}}^{\rm{i}}, {\widehat{\rm{X}}}^{\rm{i}})$

(7)

where L denotes the loss function, and $L\left(X, \widehat{X}\right) = \parallel X-\widehat{X}{\parallel }^{2}$ represents the loss function.

Now Hoeffeding tree is used to classify the class labels. The Hoeffding tree is a kind of decision tree that consists of a root node, test node and leaf node. The leaf node holds the class prediction. The main requirement in streaming data is to classify the data in a single pass. The data represented as a tree structure using the Hoeffding tree technique when the model built incrementally. The main disadvantage in Hoeffding tree classification is that it fails to classify the data into a tree when a tie occurs. Equation (8) gives the formula for Hoeffding bound calculation ^[25].

$\in = \sqrt{\frac{\left({R}^{2}\rm{log}\left(\frac{1}{\delta }\right)\right)}{2n}}$

(8)

where R denotes the range of random variable, δ denotes the desired probability not within ε of the expected value, N denotes the number of instances collected at the node.

Algorithm: Stacked Autoencoder Hoeffding Tree Approach
Input: NSL_KDD benchmark dataset
Output: Classification results: Accuracy, Sensitivity, Specificity, False Alarm Rate, False Negative Rate and F₁ Score.
Procedure
1. Load the network intrusion benchmark dataset (NSL_KDD dataset)
2. Data preprocessing
3.Apply bio-inspired feature selection technique (DPSO) to select the significant features
4. Partition the dataset into training and testing data
5. Input data to the encoder and maps it to the hidden representation to obtain a learned feature vector.
6. The feature vector from the previous layer is the input to the next layer. This process repeats till the training ends.
7. The decoder reconstructs the input from the hidden representation. Thus, Eq (7) minimizes the reconstruction error.
8. for entire training data, do
9. Use the Hoeffding tree technique and sort the instances to $f$ leaf
10. Update the necessary statistics in $f$
11. Increment ${m}_{f}$ , for all the instances at $f$
12. if ${m}_{f}$ mod ${m}_{min}=\rm{ }0$ and instances at $f$ are of a different class, then
13. Calculate ${G}_{f}k\left(i\right)$ for each feature
14. The feature with the highest ${G}_{f}$ value represents ${k}_{q}$
15. The feature with the second-highest ${G}_{f}$ value represents ${k}_{q}$
16.Calculate Hoeffding Tree Bound using Eq (8)
17. If ${k}_{q}$ ≠ ${k}_{\rm{\varnothing }}$ and ( ${G}_{f}{(k}_{p})$ - ${G}_{f}{(k}_{q})$ > ∈ or ∈< τ) then
18. Replace $f$ with an internal node that splits on ${k}_{p}$
19. for entire branches, do
20. Add and initialize a new leaf with sufficient statistics
21. End
22.Returns the classification result.

| Show Table

DownLoad: CSV

The pseudocode of the stacked autoencoder hoeffdding tree explains the following: Line 1 loads and streams the NSL_KDD dataset using Matlab platform. Line 2 performs the preprocessing of the dataset. The categorical data in the dataset is converted to numeric data using the one-hot encoding technique. In line 3, the DPSO feature selection algorithm is applied, and the important features are selected based on the selection score generated by the algorithm. Line 4 the data is divided into test and train sets. Lines 5–7 explains the stacked autoencoder process where the input data to the encoder is mapped to the hidden representations to obtain the learned features. The learned feature vector is given as an input to the next layer. This process continues till the training ends. The decoder reconstructs the hidden representations. The reconstruction error is minimized by calculating the loss function. Lines 8–21 explains the procedure of the hoeffding tree. The output generated by the stacked autoencoder is fed as an input to the hoeffding tree technique. The input is taken and the root node is decided and initialized. The tree is constructed incrementally for each training data until suitable leaf arrives. Each node has enough information to make a decision. It uses information gain to make the attribute split. The best attribute is found at each node and test is performed to decide whether the attribute yields better results based on Hoeffding bound. The test is applied on the attributes to find out which attribute gives better results and split the node for the growth of the tree. Line 22 returns the classification results.

4. Experimental setup

The experiment was conducted with the proposed technique in MATLAB R2021a on Windows 10 64-bit operating system with Ryzen 7 processor and 16 GB RAM. The experiment uses a stream-oriented offline database for querying the network traffic data. It enables a natural data analysis within the IDS. The streaming architecture is used across multiple sites to process attack data to increase the performance in large scale systems because the data is processed during the natural flow and stored only for a limited amount of time for analysis. In the research work, we have applied a hybrid classification technique. The proposed hybrid classification technique's performance was analyzed by applying various performance evaluation metrics.

4.1. Intrusion dataset

The NSL_KDD dataset is used for experimentation of the proposed work. The KDD CUP'99 dataset is the popular benchmark dataset used for network intrusion detection system. The main limitation of the KDD CUP'99 dataset is that it contains a high number of redundant records that affect the effectiveness of the evaluated system ^[26]. The improved version of KDD CUP'99 is the NSL_KDD dataset in which the redundant records are removed. NSL_KDD dataset have approximately 125, 973 training data and 22, 544 testing data ^[27]. Similar to KDD CUP'99, the records in NSL KDD dataset are unique and labelled as normal and anomaly. It has 41 features that address four different categories of attacks. Table 2 shows the NSL KDD features ^[27].

Table 2. NSL_KDD dataset features.

No	NSL_KDD Feature Names	No	NSL_KDD Feature Names
1	duration	21	_is_host_login
2	protocols_types	22	_is_guests_login
3	services	23	_counts
4	flag	24	src_counts
5	source_bytes	25	srcerror_rate
6	dstn_bytes	26	srcs_error_rate
7	lang	27	rerrors_rate
8	wrong_fragments	28	src_rerrors_rate
9	urgent	29	simlr_srcs_rate
10	hot	30	diff_srcs_rate
11	numbr_failed_login	31	src_diffs_host_rate
12	usr_login	32	dstv_host_counts
13	numbr_compromises	33	dstn_host_src_count
14	numbr_root_shell	34	dstn_simlr_src_rate
15	numbr_attempts	35	dstn_diffs_srce_rate
16	numbr_of_roots	36	dstn_hosts_sim_srce_port_rates
17	numbr_files_creation	37	dstn_hosts_src_diffr_host_rates
18	numbr_offshell	38	dstn_hosts_srcerror_rate
19	numbr_access_usrfiles	39	dstn_hosts_srce_srerror_rate
20	numbr_outbounds_cmd	40	dstn_hosts_error_rates
		41	dstn_hosts_srce_error_rates

| Show Table

DownLoad: CSV

4.2. Performance evaluation

The study was conducted to measure the performance and compare the results based on the performance metrics such as accuracy, sensitivity, specificity, False Positive Rate (FPR), False Negative Rate (FNR). The feature selection technique selects the important features that play a significant role in the intrusion. Then the proposed classifier performance is determined with the help of performance evaluation metrics. Accuracy is the most common performance evaluation technique. Accuracy is the total number of instances correctly predicted as an attack from the ratio of all predictions made from the given dataset. Equation (9) shows the way of accuracy (ACC) calculation ^[28].

$A\rm{C}\rm{C} = \frac{\rm{T}\rm{P}+\rm{T}\rm{N}}{\rm{T}\rm{P}+\rm{T}\rm{N}+\rm{F}\rm{P}+\rm{F}\rm{N}}$

(9)

The confusion matrix in Table 3 defines the TP, TN, FP, FN

Table 3. Confusion matrix for IDS.

Actual Instance	Predicted Instance
Actual Instance	Normal	Anomaly
Normal	True Negative (TN)	False Positive (FP)
Anomaly	False Negative (FN)	True Positive (TP)

| Show Table

DownLoad: CSV

TN (True Negative): The actual and predicted instances both are classified as normal.

FP (False Positive): The actual normal instance is predicted as an anomaly by the IDS.

FN (False Negative): The actual anomaly instance is predicted as normal by the IDS.

TP (True Positive): The actual anomaly instance is predicted as an anomaly by the IDS.

Sensitivity (S) is the probability to identify an attack instance as an attack accurately. Recall or True Positive Rate (TPR) are the other names for sensitivity. Sensitivity can be calculated as in Eq (10) ^[29].

$\rm{S} = \frac{\rm{T}\rm{P}}{\rm{T}\rm{P}+\rm{F}\rm{N}}$

(10)

Specificity (SP) is the probability to identify a normal instance as normal correctly. Specificity can be named as True Negative Rate (TNR). Equation (11) shows the specificity calculation formula ^[29].

$\rm{S}\rm{P} = \frac{\rm{T}\rm{N}}{\rm{T}\rm{N}+\rm{F}\rm{P}}$

(11)

FPR is the chance that the normal instance incorrectly classified as an attack. FPR is also known as False Alarm Rate (FAR). FPR calculation is given by Eq (12) ^[30].

$\rm{F}\rm{P}\rm{R} = \frac{\rm{F}\rm{P}}{\rm{F}\rm{P}+\rm{T}\rm{N}}$

(12)

FNR is the chance that the attack instance incorrectly classified as normal. It is also named Miss Rate. FNR calculation is given in Eq (13) ^[30].

$\rm{F}\rm{N}\rm{R} = \frac{\rm{F}\rm{N}}{\rm{F}\rm{N}+\rm{T}\rm{P}}$

(13)

F₁ Score is the harmonic mean of precision and recall. F₁ Score serves as the derived effectiveness measurement. F₁ Score calculation is given by Eq (14) ^[30].

${\rm{F}}_{1}\rm{S}\rm{c}\rm{o}\rm{r}\rm{e} = 2\rm{*}\frac{\rm{P}\rm{r}\rm{e}\rm{c}\rm{i}\rm{s}\rm{i}\rm{o}\rm{n}\rm{*}\rm{R}\rm{e}\rm{c}\rm{a}\rm{l}\rm{l}}{\rm{P}\rm{r}\rm{e}\rm{c}\rm{i}\rm{s}\rm{i}\rm{o}\rm{n}+\rm{R}\rm{e}\rm{c}\rm{a}\rm{l}\rm{l}}$

(14)

We designed an overall system architecture for supporting the properties of Network IDS with the help of machine learning techniques. Figure 2 shows the flow diagram of system architecture consisting of network benchmark dataset, feature selection, proposed classifier, evaluation parameters and analysis of results.

5. Results and discussion

In this experimental analysis, the feature selection technique selects the optimal subset of informative features from the given dataset. We take the optimal subset of features for analyzing the impact of the proposed technique on how accurately it classifies the network traffic into normal or anomaly.

A comprehensive study was conducted to validate the impact of the proposed technique, such as performance metrics, feature selection analysis which is helpful in the prediction of anomalies in the network traffic. The main objective of this experimental study is that.

● To identify the optimal subset of informative features that contributes to intrusion from the intrusion dataset. Though the DPSO feature selection technique is not new, the main advantage of DPSO is that it won't get trapped in the local optimum that helps us select the best features for evaluation.

● To enhance the efficiency of intrusion detection using a hybrid classification technique, which is novel and is used to evaluate the performance and detection capabilities.

● To evaluate the proposed classification technique on the NSL_KDD dataset.

● To study the performance evaluation of the proposed classification technique by applying various performance measures such as accuracy, sensitivity, specificity, FPR and FNR.

● To compare the efficiency of the proposed technique with state-of-art methods.

5.1. Feature selection techniques analysis

A comprehensive study was performed between five different bio-inspired feature section techniques in terms of accuracy, detection rate and FPR. The proposed feature selection technique achieved a considerable performance improvement compared to other feature selection techniques. The results achieved after applying the feature selection and proposed classification technique to the NSL_KDD dataset is depicted in Table 4. It compares classification accuracy, classifiers detection rate and false positive rate of the proposed technique with the state-of-art techniques. In the compared feature selection techniques LHHLS-KH feature selection technique has given the best accuracy of 96.12%, detection rate of 96.48% and false positive rate of 4%. Now the performance of the proposed technique combined with feature selection gives an accuracy of 97.70 %, detection rate of 97% and false positive rate of 1.25 % which is 0.52, 1.58 and 2.75% higher than LHHLS-KH technique.

Table 4. Classification accuracy, DR and FPR of various feature selection techniques (NSL_KDD dataset).

Feature Selection	Number of Features	Selected features	Accuracy (%)	Detection Rate (DR) (%)	False Positive Rate (FPR) (%)
CMPSO ^[7]	33	2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 17, 18, 19, 20, 21, 22, 23, 25, 26, 27, 30, 32, 33, 34, 35, 37, 40, 41	80.91	83.01	24.70
ACO ^[7]	31	1, 3, 4, 6, 7, 8, 12, 14, 15, 16, 17, 19, 20, 21, 23, 24, 25, 27, 28, 29, 30, 33, 34, 35, 36, 37, 38, 39, 40, 41	84.03	87.15	19.30
KH ^[7]	26	2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 13, 14, 15, 18, 19, 21, 22, 23, 24, 26, 28, 30, 31, 32, 40, 41	88.21	89.46	12.88
IKH ^[7]	25	2, 3, 4, 5, 6, 8, 10, 11, 12, 14, 17, 18, 20, 21, 22, 27, 28, 29, 30, 31, 34, 35, 36, 39, 41	91.22	91.75	7.34
LHHLS-KH ^[7]	19	2, 3, 4, 6, 8, 10, 11, 15, 17, 19, 20, 21, 29, 30, 33, 34, 36, 37, 40	96.12	96.48	4
DPSO	28	2, 3, 4, 6, 7, 8, 11, 12, 13, 15, 16, 17, 18, 19, 21, 23, 24, 25, 26, 28, 29, 30, 33, 34, 35, 36, 37, 40	97.70	97	1.25

| Show Table

DownLoad: CSV

To visualize the difference between classification accuracy, DR and FPR are shown in Figure 3. For the NSL_KDD dataset, the false positive rate of DPSO with the proposed classification technique is 1.25%. It reduces by 23.45, 18.05, 11.63, 6.09 and 2.75%, respectively, compared with Cross Mutation Particle Swarm Optimization (CMPSO), Ant Colony Optimization (ACO), Krill Herd (KH), Improved Krill Herd (IKH) and Linear Nearest Neighbor Lasso Step-Krill Herd (LHHLS-KH). Similarly, the detection rate of the DPSO with the proposed classification technique is 97% which is 13.99, 9.85, 7.54, 5.25, 0.52% higher than CMPSO, ACO, KH, IKH and LHHLS-KH. Similarly, the accuracy of DPSO with the proposed classification technique is 97.70% which is 16.79, 13.67, 9.49, 6.48 and 1.58% higher than CMPSO, ACO, KH, IKH and LHHLS-KH. In conclusion, the proposed classification technique with DPSO proves that it has higher detection accuracy, lower false positive rate and higher detection rate.

Figure 3. Comparison of classification accuracy, DR and FPR with FS techniques (NSL_KDD dataset).

DownLoad: Full-Size Img PowerPoint

5.2. Detection time analysis

The detection time was compared for further evaluation. Table 5 shows the detection time when applied to different techniques. The detection time represents the time taken from inputting the optimal subset of features into the classifier till the end of detection.

Table 5. Detection time of different feature selection techniques (NSL_KDD dataset).

Feature Selection Techniques	Detection Time (Sec)
CMPSO ^[20]	41.36
ACO ^[20]	41.39
KH ^[20]	37.78
IKH ^[20]	36.99
LNNLS-KH ^[20]	34.40
DPSO	41.56

| Show Table

DownLoad: CSV

Table 5 shows the time taken to detect the intrusion when different bio-inspired feature selection techniques are applied. The features selected by CMPSO, ACO, KH, IKH and LHHLS-KH are inputted to the individual classifier. In the existing systems compared, ACO feature selection took the highest detection time of 41.39 seconds with the accuracy of 84.03%, detection rate of 87.15% and false positive rate of 19.30%. Now the selected features with DPSO feature selection technique when given to a proposed hybrid classifier it took 41.56 seconds to detect the intrusion which is 0.41% higher than ACO feature selection technique. Even though it took slightly longer detection time there is a considerable increase in the performance. The DPSO detection accuracy is 97.7%, detection rate is 97% and false positive rate is 1.25% which when compared to ACO yields 13.67, 9.85% increase in accuracy, detection rate and 18.05% decrease in false positive rate.

5.3. Performance analysis of proposed classifier

The experimental study was conducted using the proposed stacked autoencoder hoeffding tree approach with and without feature selection. The proposed technique performance evaluation was based on accuracy, sensitivity, specificity, false-positive rate and false-negative rate. The results achieved with and without the feature selection of the proposed classification technique is given in Table 6.

Table 6. Performance comparison of the proposed technique with and without feature selection.

Performance Metrics (%)	SAE-HT without FS (41 Features)	SAE-HT with FS (28 Features)
Accuracy	55.93	97.7
Sensitivity (TPR)	100	97
Specificity (TNR)	28.57	98.75
False Positive Rate (FPR)	71.43	1.25
False Negative Rate (FNR)	0	3
F1 Score	37.6	98

| Show Table

DownLoad: CSV

The proposed SAE-HT was employed on the NSL_KDD dataset. Figure 4 demonstrates the performance comparison of proposed technique with and without FS. First, the proposed classifier was applied to all the 41 features of the dataset and found that the detection accuracy was only 55.93%. Moreover, the false positive rate was 71.43% and false negative rate was almost 0%. Our aim was to reduce the false positive rate to a greater extent and increase the accuracy as far as possible. Then the study was made and found that less relevant and unimportant features affect the performance of the proposed classifier. Second, the bio-inspired DPSO feature selection technique was applied and 28 important features that contributes to intrusion were selected. The selected features were given to the stacked autoencoder to learn the features so as to yield good performance and the output from the stacked autoencoder was given to hoeffding tree and found that the accuracy increased to 97.7%, false positive rate decreased to 1.25% which is 43.77% higher than SAE-HT without FS. Same way the false positive rate decreased by 70.18% when SAE-HT with FS was applied. Likewise, specificity was 97% and sensitivity rate was 98.75% when SEA-HT with FS was applied. Therefore, it proves that the proposed classifier with feature selection provides a good detection accuracy with less FPR and FNR.

Figure 4. Performance comparison of the proposed technique with and without feature selection.

DownLoad: Full-Size Img PowerPoint

The exact process is carried out for 29 features, and the accuracy is 97.3%. When applied to 30 features, the accuracy is 97.1%. From this, we conclude that the accuracy reaches saturation. There is no drastic improvement in accuracy when we include more features. As we increase the number of features once again, the accuracy starts decreasing. Therefore 28 features are the optimal features that contribute to intrusion. As the accuracy increases the false positive rate started decreasing.

Figure 5 depicts the performance of Stacked Autoencoder (SAE) and Stacked Autoencoder Hoeffding Tree (SAE-HT). The stacked autoencoder by itself gives a good accuracy, detection rate of 85.23%, 85.13% and reduced false alarm rate of 14.62% but achieves much better results in terms of accuracy, detection rate and false alarm rate when combined with the hoeffding tree. SAE-HT technique when applied it gave the accuracy of 97.7%, Detection Rate of 97% which is 12.47, 11.87% higher than SAE and false alarm rate of 1.25% which is decreased by 13.37% when compared to SAE.

Figure 5. Performance comparison: SAE Vs SAE-HT.

DownLoad: Full-Size Img PowerPoint

5.4. Comparison with state-of-art techniques

The proposed technique was compared with the state-of-Art techniques to prove its efficiency. Table 7 show the comparison of proposed with other state-of-art techniques. We have compared different conventional classification methods and deep leaning methods with our proposed technique.

Table 7. Comparison of proposed technique with state-of-art techniques.

Prediction Techniques	Accuracy (%)
VLSTM ^[8]	89.5
AdaBoost ^[8]	84.8
LSTM ^[8]	85.1
CNN LSTM ^[8]	83.3
SAE-SVM ^[10]	95.98
SAE ^[11]	85.23
SSAE ^[11]	86.02
DAE ^[11]	86.92
ContAE ^[11]	87.98
CAE ^[11]	81.07
SAE-DNN ^[12]	82.23
SAAE-DNN ^[12]	87.74
Hoeffding Tree ^[14]	93
Restricted Hoeffding Trees ^[14]	92.15
AUE2 with Buffer ^[14]	94.07
AUE2 without Buffer ^[14]	94.06
HAT + DDM + ADWIN ^[14]	92
Hierarchical Clustering and SVM ^[17]	95.72
Proposed SAE-HT	97.7

| Show Table

DownLoad: CSV

The SAE-HT classifier with feature selection gives the overall detection accuracy of 97.7 % which increases by 8.2, 12.9, 12.6, 14.4, 1.72, 12.47, 11.68, 10.78, 9.72, 16.7, 15.47, 9.96, 4.7, 5.55, 3.63, 3.64, 5.7 and 1.98% when compared to VLSTM, AdaBoost, LSTM, CNN LSTM, SAE-SVM, SAE, SAAE, DAE, Cont AE, CAE, SAE-DNN, SAAE-DNN, Hoeffding tree, Restricted hoeffding tree, AUE2 with buffer, AUE2 without buffer, HAT+DDM+ADWIN, Hierarchical Clustering and SVM.

Compared with the state-of-the-art techniques in the related work, the proposed SAE-HT classifier performs better in terms of accuracy, detection rate, and false alarm rate. Thus, proving that it has met the significant challenges by giving high accuracy of 97.7% and detection rate of 97% and well reduced false alarm rate of 1.25%.

6. Conclusions

An effective intrusion detection system was developed using the stacked autoencoder hoeffding tree technique for classification and the Darwinian particle swarm optimization method for feature selection. Using DPSO feature selection, the optimal subset of features contributing to intrusions are selected from the given intrusion dataset. The training and testing of data are given to our proposed classification technique to classify the network attacks. The main goal of the stacked autoencoder hoeffding tree technique is to increase the accuracy and reduce the false alarm rate. Moreover, we evaluated our proposed work on the NSL_KDD intrusion dataset. The dataset is streamed in a Matlab environment. Our proposed technique is applied and measures the classification performance using accuracy, specificity, sensitivity, detection rate, false alarm rate, false-negative rate. This classification performance is compared with other states of the art methods. The main challenge with the current intrusion detection is that it cannot detect the new unknown attacks. We proved that our proposed technique shows a robust significant amount of performance improvement in detection accuracy, thus reducing the false alarm rate to a greater extent. The supervised learning technique mainly depends on the predefined attack signatures that make new attacks goes undetected. The proposed stacked autoencoder hoeffding tree detects known attacks and, at the same time, detects unknown attacks to a greater extent. Results are compared with the state-of-art techniques and found that the SAE-HT technique gives higher accuracy of 97.7%, which increases by 8, 12.9, 12.6, 14.4%, 1.72, 12.47, 11.68, 10.78, 9.72, 16.7, 15.47, 9.96, 4.7, 5.55, 3.63, 3.64, 5.7 and 1.98% when compared to VLSTM, AdaBoost, LSTM, CNN LSTM, SAE-SVM, SAE, SAAE, DAE, Cont AE, CAE, SAE-DNN, SAAE-DNN, Hoeffding tree, Restricted hoeffding tree, AUE2 with buffer, AUE2 without buffer, HAT + DDM + ADWIN, Hierarchical Clustering and SVM and higher detection rate of 97%. Thus, reducing the false alarm rate to 1.25%. Further work can incorporate a study on a more effective method to improve the IDS. For the attack with fewer samples, an adversarial learning method can be used to create similar attacks, thus increasing the diversity of training examples that improve the detection accuracy.

Acknowledgments

This Publication is an outcome of the R & D work under MeitY's Visvesvaraya PhD Scheme, Government of India, implemented by Digital India Corporation.

Conflict of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Funding

This research was supported by the Faculty of Management of Comenius University in Bratislava, Slovakia.

Conflict of interest

All authors declare no conflicts of interest in this paper.

References

[1]	Lauer SA, Grantz KH, Bi Q, et al. (2020) The incubation period of coronavirus disease 2019 (COVID-19) from publicly reported confirmed cases: Estimation and application. Ann Intern Med 172: 577-582. https://doi.org/10.7326/M20-0504
[2]	Dai L, Gao GF (2021) Viral targets for vaccines against COVID-19. Nat Rev Immunol 21: 73-82. https://doi.org/10.1038/s41577-020-00480-0
[3]	Townsend JP, Hassler HB, Wang Z, et al. (2021) The durability of immunity against reinfection by SARS-CoV-2: a comparative evolutionary study. Lancet Microbe 2: e666-e675. https://doi.org/10.1016/S2666-5247(21)00219-6
[4]	Choi JY, Smith DM (2021) SARS-CoV-2 variants of concern. Yonsei Med J 62: 961-968. https://doi.org/10.3349/ymj.2021.62.11.961
[5]	Petersen E, Ntoumi F, Hui DS, et al. (2022) Emergence of new SARS-CoV-2 variant of concern omicron (B.1.1.529)—highlights Africa's research capabilities, but exposes major knowledge gaps, inequities of vaccine distribution, inadequacies in global COVID-19 response and control efforts. Int J Infect Dis 114: 268-272. https://doi.org/10.1016/j.ijid.2021.11.040
[6]	Alsaadi E, Jones IM (2019) Membrane binding proteins of coronaviruses. Future Virol 14: 275-286. https://doi.org/10.2217/fvl-2018-0144
[7]	Walls AC, Park Y, Tortorici MA, et al. (2020) Structure, function and antigenicity of the SAR-CoV-2 spike glycoprotein. Cell 180: 281-292. https://doi.org/10.1016/j.cell.2020.02.058
[8]	Chukwudozie OS, Gray CM, Fagbayi TA, et al. (2021) Immuno-informatics design of a multimeric epitope peptide based vaccine targeting SARS-CoV-2 spike glycoprotein. PloS One 16: e0248061. https://doi.org/10.1371/journal.pone.0248061
[9]	Amanat F, Krammer F (2020) SARS-CoV-2 vaccines: Status report. Immunity 52: 583-589. https://doi.org/10.1016/j.immuni.2020.03.007
[10]	Tosta E (2021) The protective immunity induced by SARS-CoV-2 infection and vaccination: a critical appraisal. Explor Immunol 1: 199-225. https://doi.org/10.37349/ei.2021.00014
[11]	Sumon TA, Hussain MA, Hasan MT, et al. (2021) A revisit to the research updates of drugs, vaccines, and bioinformatics approaches in combating COVID-19 pandemic. Front Mol Biosci 7: 585899. https://doi.org/10.3389/fmolb.2020.585899
[12]	Garg M, Maralakunte M, Kumar Y, et al. (2021) Vaccine-induced immune responses against SARS-CoV-2 infections. Explor Immunol 1: 356-373. https://doi.org/10.37349/ei.2021.00024
[13]	Sonani B, Aslam F, Goyal A, et al. (2021) COVID-19 vaccination in immunocompromised patients. Clin Rheumatol 40: 797-798. https://doi.org/10.1007/s10067-020-05547-w
[14]	Pollard AJ, Bijker EM (2021) Publisher correction: A guide to vaccinology: from basic principles to new developments. Nat Rev Immunol 21: 129. https://doi.org/10.1038/s41577-020-00497-5
[15]	Peng Y, Mentzer AJ, Liu G, et al. (2020) Broad and strong memory CD4+ and CD8+ T cells induced by SARS-CoV-2 in UK convalescent individuals following COVID-19. Nat Immunol 21: 1336-1345. https://doi.org/10.1038/s41590-020-0782-6
[16]	Kim YI, Casel MAB, Choi YK (2022) Transmissibility and pathogenicity of SARS-CoV-2 variants in animal models. J Microbiol 60: 255-267. https://doi.org/10.1007/s12275-022-2033-z
[17]	Chemaitelly H, Yassine HM, Benslimane FM, et al. (2021) mRNA-1273 COVID-19 vaccine effectiveness against the B.1.1.7 and B.1.351 variants and severe COVID-19 disease in Qatar. Nat Med 27: 1614-1621. https://doi.org/10.1038/s41591-021-01446-y
[18]	(2021) WHOTracking SARS-CoV-2 variants. World Health Organization. Available from: https://www.who.int/en/activities/tracking-SARS-CoV-2-variants/
[19]	Chaillon A, Smith DM, et al. (2021) Phylogenetic analyses of severe acute respiratory syndrome Coronavirus 2 (SARS-CoV-2) B.1.1.7 lineage suggest a single origin followed by multiple exportation events versus convergent evolution. Clin Infect Dis 73: 2314-2317. https://doi.org/10.1093/cid/ciab265
[20]	Davies NG, Abbott S, Barnard RC, et al. (2021) Estimated transmissibility and impact of SARS-CoV-2 lineage B.1.1.7 in England. Science 372: eabg3055. https://doi.org/10.1126/science.abg3055
[21]	Subissi L, von Gottberg A, Thukral L, et al. (2022) An early warning system for emerging SARS-CoV-2 variants. Nat Med 28: 1110-1115. https://doi.org/10.1126/science.abg3055
[22]	Graham MS, Sudre CH, May A, et al. (2021) Changes in symptomatology, reinfection, and transmissibility associated with the SARS-CoV-2 variant B.1.1.7: an ecological study. Lancet Public Health 6: e335-e345. https://doi.org/10.1016/s2468-2667(21)00055-4
[23]	Chavda VP, Patel AB, Vaghasiya DD, et al. (2022) SARS-CoV-2 variants and vulnerability at the global level. J Med Virol 94: 2986-3005. https://doi.org/10.1002/jmv.27717
[24]	Meng B, Kemp SA, Papa G, et al. (2021) Recurrent emergence of SARS-CoV-2 spike deletion H69/V70 and its role in the Alpha variant B.1.1.7. Cell Rep 35: 109292. https://doi.org/10.1016/j.celrep.2021.109292
[25]	Starr TN, Greaney AJ, Hilton SK, et al. (2020) Deep mutational scanning of SARS-CoV-2 receptor binding domain reveals constraints on folding and ACE2 binding. Cell 182: 1295-1310. https://doi.org/10.1016/j.cell.2020.08.012
[26]	Collier DA, De Marco A, Ferreira IA, et al. (2022) Author correction: Sensitivity of SARS-CoV-2 B.1.1.7 to mRNA vaccine-elicited antibodies. Nature 608: E24. https://doi.org/10.1038/s41586-022-05103-3
[27]	Dagan N, Barda N, Kepten E, et al. (2021) BNT162b2 mRNA COVID-19 vaccine in a nationwide mass vaccination setting. N Engl J Med 384: 1412-1423. https://doi.org/10.1056/nejmoa2101765
[28]	Barbra AD, Gerlovin H, Madenci AL, et al. (2022) Comparative effectiveness of BNT162b2 and mRNA-1273 vaccines in U.S. veterans. N Engl J Med 386: 105-115. https://doi.org/10.1056/nejmoa2115463
[29]	McNamara LA, Wiegand RE, Burke RM, et al. (2022) Estimating the early impact of the US COVID-19 vaccination programme on COVID-19 cases, emergency department visits, hospital admissions, and deaths among adults aged 65 years and older: an ecological analysis of national surveillance data. Lancet 399: 152-160. https://doi.org/10.1016/s0140-6736(21)02226-1
[30]	George NI, Locke ER, O'Hare AM, et al. (2022) COVID-19 vaccination effectiveness against infection or death in a national US health care system: A target trial emulation study. Ann Intern Med 175: 352-361. https://doi.org/10.7326/m21-3256
[31]	Mahase E (2021) Covid-19: Novavax vaccine efficacy is 86% against UK variant and 60% against South African variant. BMJ 372: n296. https://doi.org/10.1136/bmj.n296
[32]	Mark WT, Self WH, Gaglani M, et al. (2022) Effectiveness of mRNA vaccination in preventing COVID-19-associated invasive mechanical ventilation and death—United States, March 2021–January 2022. MMWR Morb Mortal Wkly Rep 71: 459-465. https://doi.org/10.15585/mmwr.mm7112e1
[33]	Khan A, Zia T, Suleman M, et al. (2021) Higher infectivity of the SARS-CoV-2 new variants is associated with K417N/T, E484K, and N501Y mutants: An insight from structural data. J Cell Physiol 236: 7045-7057. https://doi.org/10.1002/jcp.30367
[34]	William NW, Ip S, Cooper JA, et al. (2022) Association of COVID-19 vaccines ChAdOx1 and BNT162b2 with major venous, arterial, or thrombocytopenic events: A population-based cohort study of 46 million adults in England. PLoS Med 19: e1003926. https://doi.org/10.1371/journal.pmed.1003926
[35]	Andeweg SP, Vennema H, Veldhuijzen I, et al. (2023) Elevated risk of infection with SARS-CoV-2 Beta, Gamma, and Delta variants compared with Alpha variant in vaccinated individuals. Sci Transl Med 15: eabn4338. https://doi.org/10.1126/scitranslmed.abn4338
[36]	Boehm E, Kronig I, Neher RA, et al. (2021) Novel SARS-CoV-2 variants: the pandemics within the pandemic. Clin Microbiol Infect 27: 1109-1117. https://doi.org/10.1016%2Fj.cmi.2021.05.022
[37]	Kustin T, Harel N, Finkel U, et al. (2021) Evidence for increased breakthrough rates of SARS-CoV-2 variants of concern in BNT162b2-mRNA-vaccinated individuals. Nat Med 27: 1379-1384. https://doi.org/10.1038/s41591-021-01413-7
[38]	Liu J, Liu Y, Xia H, et al. (2021) BNT162b2-elicited neutralization of B.1.617 and other SARS-CoV-2 variants. Nature 596: 273-275. https://doi.org/10.1038/s41586-021-03693-y
[39]	Xie X, Liu Y, Liu J, et al. (2021) Neutralization of SARS-CoV-2 spike 69/70 deletion, E484K and N501Y variants by BNT162b2 vaccine-elicited sera. Nat Med 27: 620-621. https://doi.org/10.1038/s41591-021-01270-4
[40]	Singh H, Dahiya N, Yadav M, et al. (2022) Emergence of SARS-CoV-2 new variants and their clinical significance. Can J Infect Dis Med Microbiol 2022: 7336309. https://doi.org/10.1155/2022%2F7336309
[41]	(2021) FDAFact sheet for health care providers emergency use authorization (EUA) of bamlanivimab and etesevimab. US Food and Drug Administration. Available from: https://www.fda.gov/media/145802/download
[42]	Wang Z, Schmidt F, Weisblum Y, et al. (2021) mRNA vaccine-elicited antibodies to SARS-CoV-2 and circulating variants. Nature 592: 616-622. https://doi.org/10.1038/s41586-021-03324-6
[43]	Prete CA, Buss LF, Buccheri R, et al. (2022) Reinfection by the SARS-CoV-2 Gamma variant in blood donors in Manaus, Brazil. BMC Infect Dis 22: 127. https://doi.org/10.1186/s12879-022-07094-y
[44]	Banho CA, Sacchetto L, Campos GRF, et al. (2022) Impact of SARS-CoV-2 Gamma lineage introduction and COVID-19 vaccination on the epidemiological landscape of a Brazilian city. Commun Med 2: 41. https://doi.org/10.1038/s43856-022-00108-5
[45]	Buss LF, Prete CA, Abrahim CMM, et al. (2021) Three-quarters attack rate of SARS-CoV-2 in the Brazilian Amazon during a largely unmitigated epidemic. Science 371: 288-292. https://doi.org/10.1126/science.abe9728
[46]	Sabino EC, Buss LF, Carvalho MPS, et al. (2021) Resurgence of COVID-19 in Manaus, Brazil, despite high seroprevalence. Lancet 397: 452-455. https://doi.org/10.1016/s0140-6736(21)00183-5
[47]	Volz E, Mishra S, Chand M, et al. (2021) Assessing transmissibility of SARS-CoV-2 lineage B.1.1.7 in England. Nature 593: 266-269. https://doi.org/10.1038/s41586-021-03470-x
[48]	Lucas C, Vogels CBF, Yildirim I, et al. (2021) Impact of circulating SARS-CoV-2 variants on mRNA vaccine-induced immunity. Nature 600: 523-529. https://doi.org/10.1038/s41586-021-04085-y
[49]	Collier DA, Ferreira IATM, Kotagiri P, et al. (2021) Age-related immune response heterogeneity to SARS-CoV-2 vaccine BNT162b2. Nature 596: 417-422. https://doi.org/10.1038/s41586-021-03739-1
[50]	Wang P, Casner RG, Nair MS, et al. (2021) Increased resistance of SARS-CoV-2 variant P.1 to antibody neutralization. Cell Host Microbe 29: 747-751. https://doi.org/10.1101/2021.03.01.433466
[51]	Firouzabadi N, Ghasemiyeh P, Moradishooli F, et al. (2023) Update on the effectiveness of COVID-19 vaccines on different variants of SARS-CoV-2. Int Immunopharmacol 117: 109968. https://doi.org/10.1016/j.intimp.2023.109968
[52]	(2021) WHOClassification of omicron (B.1.1.529): SARS-CoV-2 variant of concern. World Health Organization. Available from: https://www.who.int/news/item/26-11-2021-classification-of-omicron-(b.1.1.529)-sars-cov-2-variant-of-concern
[53]	Lauring AS, Malani PN (2021) Variants of SARS-CoV-2. JAMA 326: 14181. https://doi.org/10.1001/jama.2021.14181
[54]	Alkhatib M, Svicher V, Salpini R, et al. (2021) SARS-CoV-2 variants and their relevant mutational profiles: Update summer 2021. Microbiol Spectr 9: e0109621. https://doi.org/10.1128/spectrum.01096-21
[55]	Cascella M, Rajnik M, Aleem A, et al. (2023) Features, evaluation, and treatment of coronavirus (COVID-19). StatPearls . Tampa: StatPearls Publishing.
[56]	Harvey WT, Carabelli AM, Jackson B, et al. (2021) SARS-CoV-2 variants, spike mutations and immune escape. Nat Rev Microbiol 19: 409-424. https://doi.org/10.1038/s41579-021-00573-0
[57]	Fisman DN, Tuite AR (2021) Progressive increase in virulence of novel SARS-CoV-2 variants in Ontario, Canada. CMAJ 193: E1619-E1625. https://doi.org/10.1503/cmaj.211248
[58]	King KL, Wilson S, Napolitano JM, et al. (2022) SARS-CoV-2 variants of concern Alpha and Delta show increased viral load in saliva. PloS One 17: e0267750. http://dx.doi.org/10.1371/journal.pone.0267750
[59]	(2021) CDC, SARS-CoV-2 variant classifications and definitions. Centers for Disease Control and Prevention. Available from: https://www.cdc.gov/coronavirus/2019-ncov/variants/variant-classifications.html
[60]	Kannan SR, Spratt AN, Cohen AR, et al. (2021) Evolutionary analysis of the Delta and Delta Plus variants of the SARS-CoV-2 viruses. J Autoimmun 124: 102715. https://doi.org/10.1016/j.jaut.2021.102715
[61]	Planas D, Veyer D, Baidaliuk A, et al. (2021) Reduced sensitivity of SARS-CoV-2 variant delta to antibody neutralization. Nature 596: 276-280. https://doi.org/10.1038/s41586-021-03777-9
[62]	Tchesnokova V, Kulasekara H, Larson L, et al. (2021) Acquisition of the L452R mutation in the ACE2-binding interface of spike protein triggers recent massive expansion of SARS-CoV-2 variants. J Clin Microbiol 59: e0092121. https://doi.org/10.1128/jcm.00921-21
[63]	Sheikh A, Mcmenamin J, Taylor B, et al. (2021) SARS-CoV-2 Delta VOC in Scotland: demographics, risk of hospital admission, and vaccine effectiveness. Lancet 397: 2461-2462. https://doi.org/10.1016/s0140-6736(21)01358-1
[64]	Bernal JL, Andrews N, Gower C, et al. (2021) Effectiveness of COVID-19 vaccines against the B.1.617.2 (delta) variant. N Engl J Med 385: 585-594. https://doi.org/10.1056/nejmoa2108891
[65]	Mohsin M, Mahmud S (2022) Omicron SARS-CoV-2 variant of concern: A review on its transmissibility, immune evasion, reinfection, and severity. Medicine 101: e29165. https://doi.org/10.1097/md.0000000000029165
[66]	Ntoumi FPE, Hui DS, Abubakar A, et al. (2022) Emergence of new SARS-CoV-2 variant of concern omicron (B.1.1.529)—highlights Africa's research capabilities, but exposes major knowledge gaps, inequities of vaccine distribution, inadequacies in global COVID-19 response and control efforts. Int J Infect Dis 114: 268-272. https://doi.org/10.1016/j.ijid.2021.11.040
[67]	Rio C, Malani PN (2022) COVID-19 in 2022—The beginning of the end or the end of the beginning?. JAMA 327: 2389-2390. https://doi.org/10.1001/jama.2022.9655
[68]	Yamasoba D, Kimura I, Nasser H, et al. (2022) Virological characteristics of the SARS-CoV-2 Omicron BA.2 spike. Cell 185: 2103-2115. https://doi.org/10.1016/j.cell.2022.04.035
[69]	Lassaunière R, Polacek C, Frische A, et al. (2022) Neutralizing antibodies against the SARS-CoV-2 omicron variant (BA.1) 1 to 18 weeks after the second and third doses of the BNT162b2 mRNA vaccine. JAMA Network Open 5: e2212073. https://doi.org/10.1001/jamanetworkopen.2022.12073
[70]	Cao Y, Yisimayi A, Jian F, et al. (2022) BA.2.12.1, BA.4 and BA.5 escape antibodies elicited by Omicron infection. Nature 608: 593-602. https://doi.org/10.1038/s41586-022-04980-y
[71]	Centers for Disease Control and Prevention.SARS-CoV-2 B.1.1.529 (Omicron) Variant United States. Morb Mortal Wkly Rep (2021) 70: 1731-1734. http://dx.doi.org/10.15585/mmwr.mm7050e1
[72]	Wolter N, Jassat W, Walaza S, et al. (2021) Early assessment of the clinical severity of the SARS-CoV-2 omicron variant in South Africa. Lancet 399: 437-446. https://doi.org/10.1016/s0140-6736(22)00017-4
[73]	Tegally H, Moir M, Everatt J, et al. (2022) Emergence of SARS-CoV-2 omicron lineages BA.4 and BA.5 in South Africa. Nat Med 28: 1785-1790. https://doi.org/10.1038/s41591-022-01911-2
[74]	Callaway E (2021) Heavily mutated omicron variant puts scientists on alert. Nature 600: 21. https://doi.org/10.1038/d41586-021-03552-w
[75]	Liu L, Iketani S, Guo Y, et al. (2022) Striking antibody evasion manifested by the omicron variant of SARS-CoV-2. Nature 602: 676-681. https://doi.org/10.1038/s41586-021-04388-0
[76]	Hoffmann M, Krüger N, Schulz S, et al. (2022) The Omicron variant is highly resistant against antibody-mediated neutralization: Implications for control of the COVID-19 pandemic. Cell 185: 447-456.e11. https://doi.org/10.1016/j.cell.2021.12.032
[77]	Jiang S, Hillyer C, Du L, et al. (2020) Neutralizing antibodies against SARS-CoV-2 and other human coronaviruses. Trends Immunol 41: 355-359. https://doi.org/10.1016/j.it.2020.03.007
[78]	Watanabe Y, Allen JD, Wrapp D, et al. (2020) Site-specific glycan analysis of the SARS-CoV-2 spike. Science 369: 330-333. https://doi.org/10.1126/science.abb9983
[79]	Zhou H, Ni WJ, Huang W, et al. (2022) Advances in pathogenesis, progression, potential targets and targeted therapeutic strategies in SARS-CoV-2-induced COVID-19. Front Immunol 13: 834942. https://doi.org/10.3389/fimmu.2022.834942
[80]	De Abajo FJ, Rodríguez-Martín S, Lerma V, et al. (2020) Use of renin-angiotensin-aldosterone system inhibitors and risk of COVID-19requiring admission to hospital: a case-population study. Lancet 395: 1705-1714. https://doi.org/10.1016/s0140-6736(20)31030-8
[81]	Hoffmann M, Kleine-Weber H, Schroeder S, et al. (2020) SARS-CoV-2 cell entry depends on ACE2 and TMPRSS2 and is blocked by a clinically proven protease inhibitor. Cell 181: 271-280. https://doi.org/10.1016/j.cell.2020.02.052
[82]	Xu H, Zhong L, Deng J, et al. (2020) High expression of ACE2 receptor of 2019-nCoV on the epithelial cells of oral mucosa. Int J Oral Sci 12: 8. https://doi.org/10.1038/s41368-020-0074-x
[83]	Lu S, Ye Q, Singh D, et al. (2021) The SARS-CoV-2 nucleocapsid phosphoprotein forms mutually exclusive condensates with RNA and the membrane-associated M protein. Nat Commun 12: 502. https://doi.org/10.1038/s41467-020-20768-y
[84]	Mandala VS, McKay MJ, Shcherbakov AA, et al. (2020) Structure and drug binding of the SARS-CoV-2 envelope protein transmembrane domain in lipid bilayers. Nat Struct Mol Biol 27: 1202-1208. https://doi.org/10.1038/s41594-020-00536-8
[85]	Peng Y, Du N, Lei Y, et al. (2020) Structures of the SARS-CoV-2 nucleocapsid and their perspectives for drug design. EMBO J 39: e105938. https://doi.org/10.15252/embj.2020105938
[86]	Ricci D, Etna MP, Rizzo F, et al. (2021) Innate immune response to SARS-CoV-2 infection: From cells to soluble mediators. Int J Mol Sci 22: 7017. https://doi.org/10.3390/ijms22137017
[87]	Schultze JL, Aschenbrenner AC (2021) COVID-19 and the human innate immune system. Cell 184: 1671-1692. https://doi.org/10.1016/j.cell.2021.02.029
[88]	Gu W, Gan H, Ma Y, et al. (2022) The molecular mechanism of SARS-CoV-2 evading host antiviral innate immunity. Virol J 19: 49. https://doi.org/10.1186/s12985-022-01783-5
[89]	Wauters E, Van Mol P, Garg AD, et al. (2021) Discriminating mild from critical COVID-19 by innate and adaptive immune single-cell profiling of bronchoalveolar lavages. Cell Res 31: 272-290. https://doi.org/10.1038/s41422-020-00455-9
[90]	Zhang S, Wang L, Cheng G, et al. (2022) The battle between host and SARS-CoV-2: Innate immunity and viral evasion strategies. Mol Ther 30: 1869-1884. https://doi.org/10.1016/j.ymthe.2022.02.014
[91]	Paludan SR, Mogensen TH (2022) Innate immunological pathways in COVID-19 pathogenesis. Sci Immunol 7: eabm5505. https://doi.org/10.1126/sciimmunol.abm5505
[92]	Thorne LG, Reuschl AK, Zuliani-Alvarez L, et al. (2021) SARS-CoV-2 sensing by RIG-I and MDA5 links epithelial infection to macrophage inflammation. EMBO J 40: e107826. https://doi.org/10.15252/embj.2021107826
[93]	Cheemarla NR, Watkins TA, Mihaylova VT, et al. (2021) Dynamic innate immune response determines susceptibility to SARS-CoV-2 infection and early replication kinetics. J Exp Med 218: e20210583. https://doi.org/10.1084/jem.20210583
[94]	Shah VK, Firmal P, Alam A, et al. (2020) Overview of immune response during SARS-CoV-2 infection: Lessons from the past. Front Immunol 11: 1949. https://doi.org/10.3389/fimmu.2020.01949
[95]	Li H, Liu L, Zhang D, et al. (2020) SARS-CoV-2 and viral sepsis: observations and hypotheses. Lancet 395: 1517-1520. https://doi.org/10.1016/S0140-6736(20)30920-X
[96]	Mehta P, McAuley DF, Brown M, et al. (2020) COVID-19: consider cytokine storm syndromes and immunosuppression. Lancet 395: 1033-1034. https://doi.org/10.1016/S0140-6736(20)30628-0
[97]	Karki R, Sharma BR, Tuladhar S, et al. (2021) Synergism of TNF-α and IFN-g triggers inflammatory cell death, tissue damage, and mortality in SARS-CoV-2 infection and cytokine shock syndromes. Cell 184: 149-168. https://doi.org/10.1016/j.cell.2020.11.025
[98]	Leisman DE, Ronner L, Pinotti R, et al. (2020) Cytokine elevation in severe and critical COVID-19: a rapid systematic review, meta-analysis, and comparison with other inflammatory syndromes. Lancet Respir Med 8: 1233-1244. https://doi.org/10.1016/s2213-2600(20)30404-5
[99]	Contoli M, Papi A, Tomassetti L, et al. (2021) Blood interferon-α levels and severity, outcomes, and inflammatory profiles in hospitalized COVID-19 patients. Front Immunol 12: 648004. https://doi.org/10.3389/fimmu.2021.648004
[100]	Chen G, Wu D, Guo W, et al. (2020) Clinical and immunological features of severe and moderate coronavirus disease 2019. J Clin Invest 130: 2620-2629. https://doi.org/10.1172/jci137244
[101]	Lau SKP, Lau CCY, Chan KH, et al. (2013) Delayed induction of proinflammatory cytokines and suppression of innate antiviral response by the novel Middle East respiratory syndrome coronavirus: implications for pathogenesis and treatment. J Gen Virol 94: 2679-2690. https://doi.org/10.1099/vir.0.055533-0
[102]	Pereda R, González D, Rivero HB, et al. (2020) Therapeutic effectiveness of interferon alpha 2b treatment for COVID-19 patient recovery. J Interferon Cytokine Res 40: 578-588. https://doi.org/10.1089/jir.2020.0188
[103]	Yang Y, Shen C, Li J, et al. (2020) Plasma IP-10 and MCP-3 levels are highly associated with disease severity and predict the progression of COVID-19. J Allergy Clin Immunol 146: 119-127. https://doi.org/10.1016/j.jaci.2020.04.027
[104]	Schreiber G (2020) The role of type I interferons in the pathogenesis and treatment of COVID-19. Front Immunol 11: 595739. https://doi.org/10.3389/fimmu.2020.595739
[105]	Sui L, Zhao Y, Wang W, et al. (2021) SARS-CoV-2 membrane protein inhibits type I interferon production through ubiquitin-mediated degradation of TBK1. Front Immunol 12: 662989. https://doi.org/10.3389/fimmu.2021.662989
[106]	Li A, Zhao K, Zhang B, Hua R, et al. (2021) SARS-CoV-2 NSP12 protein is not an interferon-β antagonist. J Virol 95: e0074721. https://doi.org/10.1128/jvi.00747-21
[107]	Konno Y, Kimura I, Uriu K, et al. (2020) SARS-CoV-2 ORF3b is a potent interferon antagonist whose activity is increased by a naturally occurring elongation variant. Cell Rep 32: 108185. https://doi.org/10.1016/j.celrep.2020.108185
[108]	Geng H, Subramanian S, Wu L, et al. (2021) SARS-CoV-2 ORF8 forms intracellular aggregates and inhibits IFNγ-induced antiviral gene expression in human lung epithelial cells. Front Immunol 12: 679482. https://doi.org/10.3389/fimmu.2021.679482
[109]	Wang W, Zhou Z, Xiao X, et al. (2021) SARS-CoV-2 nsp12 attenuates type I interferon production by inhibiting IRF3 nuclear translocation. Cell Mol Immunol 18: 945-953. https://doi.org/10.1038/s41423-020-00619-y
[110]	Lei J, Hilgenfeld R (2017) RNA-virus proteases counteracting host innate immunity. FEBS Lett 591: 3190-3210. https://doi.org/10.1002/1873-3468.12827
[111]	Prescott L (2022) SARS-CoV-2 3CLpro whole human proteome cleavage prediction and enrichment/depletion analysis. Comput Biol Chem 98: 107671. https://doi.org/10.1016/j.compbiolchem.2022.107671
[112]	Wang D, Fang L, Shi Y (2016) Porcine epidemic diarrhea virus 3C-like protease regulates its interferon antagonism by cleaving NEMO. J Virol 90: 2090-2101. https://doi.org/10.1128/jvi.02514-15
[113]	Gaglia MM, Covarrubias S, Wong W, et al. (2012) A common strategy for host RNA degradation by divergent viruses. J Virol 86: 9527-9530. https://doi.org/10.1128/jvi.01230-12
[114]	Zhang K, Miorin L, Makio T, et al. (2021) Nsp1 protein of SARS-CoV-2 disrupts the mRNA export machinery to inhibit host gene expression. Sci Adv 7: eabe7386. https://doi.org/10.1126/sciadv.abe7386
[115]	Finkel Y, Gluck A, Nachshon A, et al. (2021) SARS-CoV-2 uses a multipronged strategy to impede host protein synthesis. Nature 594: 240-245. https://doi.org/10.1038/s41586-021-03610-3
[116]	Lo MW, Woodruff TM (2020) Complement: Bridging the innate and adaptive immune systems in sterile inflammation. J Leukoc Biol 108: 339-351. https://doi.org/10.1002/JLB.3MIR0220-270R
[117]	Hsu JCC, Laurent-Rolle M, Pawlak JB, et al. (2021) Translational shutdown and evasion of the innate immune response by SARS-CoV-2 NSP14 protein. Proc Natl Acad Sci USA 118: e2101161118. https://doi.org/10.1073/pnas.2101161118
[118]	Noris M, Remuzzi G (2013) Overview of complement activation and regulation. Sem Nephr 33: 479-492. https://doi.org/10.1016/j.semnephrol.2013.08.001
[119]	Ali YM, Ferrari M, Lynch NJ, et al. (2021) Lectin pathway mediates complement activation by SARS-CoV-2 proteins. Front Immunol 12: 714511. https://doi.org/10.3389/fimmu.2021.714511
[120]	Holter JC, Pischke SE, de Boer E, et al. (2020) Systemic complement activation isassociated with respiratory failure in COVID-19 hospitalized patients. Proc Natl Acad Sci USA 117: 25018-25025. https://doi.org/10.1073/pnas.2010540117
[121]	Gao T, Hu M, Zhang X, et al. (2020) Highly pathogenic coronavirus N protein aggravates lung injury by MASP-2-mediated complement over-activation. medRxiv . Preprint. http://dx.doi.org/10.1101/2020.03.29.20041962
[122]	Yu J, Yuan X, Chen H, et al. (2020) Direct activation of the alternative complement pathway by SARS-CoV-2 spike proteins is blocked by factor D inhibition. Blood 136: 2080-2089. https://doi.org/10.1182/blood.2020008248
[123]	Wen W, Chen C, Tang J, et al. (2022b) Efficacy and safety of three new oral antiviral treatment (molnupiravir, fluvoxamine and Paxlovid) for COVID-19: a meta-analysis. Ann Med 54: 516-523. https://doi.org/10.1080/07853890.2022.2034936
[124]	Bernal AJ, da Silva MMG, Musungaie DB, et al. (2022) Molnupiravir for oral treatment of Covid-19 in nonhospitalized patients. N Engl J Med 386: 509-520. https://doi.org/10.1056/nejmoa2116044
[125]	Gandhi RT, Malani PN, Del Rio C (2022) COVID-19 therapeutics for nonhospitalized patients. JAMA 327: 617-618. https://doi.org/10.1001/jama.2022.0335
[126]	(2021) Pfizer Inc.Pfizer's novel COVID-19 oral antiviral treatment candidate reduced risk of hospitalization or death by 89% in interim analysis of phase 2/3 EPIC-HR study. Pfizer Inc.. Available from: https://www.pfizer.com/news/press-release/press-release-detail/pfizers-novel-covid-19-oral-antiviral-treatment-candidate
[127]	Mahase E (2021b) Covid-19: Pfizer's paxlovid is 89% effective in patients at risk of serious illness, company reports. BMJ 375: n2713. https://doi.org/10.1136/bmj.n2713
[128]	Seftel D, Boulware DR (2021) Prospective cohort of fluvoxamine for early treatment of Coronavirus disease 19. Open Forum Infect Dis 8: ofab050. https://doi.org/10.1093/ofid/ofab050
[129]	Calusic M, Marcec R, Luksa L, et al. (2022) Safety and efficacy of fluvoxamine in COVID-19 ICU patients: An open label, prospective cohort trial with matched controls. Br J Clin Pharmacol 88: 2065-2073. https://doi.org/10.1111/bcp.15126
[130]	Gottlieb RL, Vaca CE, Paredes R, et al. (2022) Early Remdesivir to Prevent Progression to Severe Covid-19 in Outpatients. N Engl J Med 386: 305-315. https://doi.org/10.1056/nejmoa2116846
[131]	Jason DG, Lye DCB, Hui DS, et al. (2020) Remdesivir for 5 or 10 days in patients with severe COVID-19. N Engl J Med 383: 1827-1837. https://doi.org/10.1056/nejmoa2015301
[132]	Spinner CD, Gottlieb RL, Criner GJ, et al. (2020) Effect of remdesivir vs standard care on clinical status at 11 days in patients with moderate COVID-19: A randomized clinical trial. JAMA 324: 1048-1057. https://doi.org/10.1001/jama.2020.16349
[133]	Vegivinti CTR, Evanson KW, Lyons H, et al. (2022) Efficacy of antiviral therapies for COVID-19: a systematic review of randomized controlled trials. BMC Infect Dis 22: 107. https://doi.org/10.1186/s12879-022-07068-0
[134]	Wang L, Zhou T, Zhang Y, et al. (2021) Antibodies with potent and broad neutralizing activity against antigenically diverse and highly transmissible SARS-CoV-2 variants. BioRxiv . Preprint. https://doi.org/10.1101/2021.02.25.432969
[135]	Wang Q, Bowen A, Valdez R, et al. (2023) Antibody response to omicron BA.4-BA.5 bivalent booster. N Engl J Med 388: 567-569. https://doi.org/10.1056/NEJMc2213907
[136]	Eric JH, Angulo FJ, McLaughlin JM, et al. (2021) Impact and effectiveness of mRNA BNT162b2 vaccine against SARS-CoV-2 infections and COVID-19 cases, hospitalisations, and deaths following a nationwide vaccination campaign in Israel: an observational study using national surveillance data. Lancet 397: 1819-1829. https://doi.org/10.1016/s0140-6736(21)00947-8
[137]	Thompson MG, Stenehjem E, Grannis S, et al. (2021) Effectiveness of Covid-19 vaccines in ambulatory and inpatient care settings. N Engl J Med 385: 1355-1371. https://doi.org/10.1056/NEJMoa2110362
[138]	Vasileiou E, Simpson CR, Shi T, et al. (2021) Interim findings from first-dose mass COVID-19 vaccination roll-out and COVID-19 hospital admissions in Scotland: a national prospective cohort study. Lancet 397: 1646-1657. https://doi.org/10.1016/s0140-6736(21)00677-2
[139]	Al Kaabi N, Zhang Y, Xia S, et al. (2021) Effect of 2 inactivated SARS-CoV-2 vaccines on symptomatic COVID-19 infection in adults: A randomized clinical trial: A randomized clinical trial. JAMA 326: 35-45. https://doi.org/10.1001/jama.2021.8565
[140]	Offit PA (2023) Bivalent COVID-19 vaccines—A cautionary tale. N Engl J Med 388: 481-483. https://doi.org/10.1056/nejmp2215780
[141]	Link-Gelles R, Ciesla AA, Fleming-Dutra KE, et al. (2022) Effectiveness of bivalent mRNA vaccines in preventing symptomatic SARS-CoV-2 infection—Increasing Community Access to testing program, United States, September-November 2022. MMWR Morb Mortal Wkly Rep 71: 1526-1530. http://dx.doi.org/10.15585/mmwr.mm7148e1
[142]	Zeng B, Gao L, Zhou Q, et al. (2022) Effectiveness of COVID-19 vaccines against SARS-CoV-2 variants of concern: a systematic review and meta-analysis. BMC Med 20: 200. https://doi.org/10.1186/s12916-022-02397-y
[143]	Fiolet T, Kherabi Y, MacDonald CJ, et al. (2022) Comparing COVID-19 vaccines fortheir characteristics, efficacy and effectiveness against SARS-CoV-2 and variants of concern: a narrative review. Clin Microbiol Infect 28: 202-221. https://doi.org/10.1016/j.cmi.2021.10.005
[144]	Graña C, Ghosn L, Evrenoglou T, et al. (2022) Efficacy and safety of COVID-19 vaccines. Cochrane Database Syst Rev 12: CD015477. https://doi.org/10.1002/14651858.CD015477
[145]	Scovino AM, Dahab EC, Vieira GF, et al. (2022) SARS-CoV-2's variants of concern: A brief characterization. Front Immunol 13: 834098. https://doi.org/10.3389/fimmu.2022.834098
[146]	Xu S, Sun M (2022) COVID-19 vaccine effectiveness during Omicron BA.2 pandemic in Shanghai: A cross-sectional study based on EMR. Medicine 101: e31763. https://doi.org/10.1097/MD.0000000000031763
[147]	Shkoda AS, Gushchin VA, Ogarkova DA, et al. (2022) Sputnik V effectiveness against hospitalization with COVID-19 during Omicron dominance. Vaccines 10: 938. https://doi.org/10.3390/vaccines10060938
[148]	Zheng C, Shao W, Chen X, et al. (2022) Effectiveness of COVID-19 vaccines against SARS-CoV-2 variants of concern in real-world: a literature review and meta-analysis. Int J Infect Dis 11: 2383-2392. https://doi.org/10.1016/j.ijid.2021.11.009
[149]	Cerqueira-Silva T, Andrews JR, Boaventura VS, et al. (2022) Effectiveness of CoronaVac, ChAdOx1 nCoV-19, BNT162b2, and Ad26.COV2.S among individuals with previous SARS-CoV-2 infection in Brazil: a test-negative, case-control study. Lancet Infect Dis 22: 791-801. https://doi.org/10.1016/S1473-3099(22)00140-2
[150]	Negahdaripour M, Shafiekhani M, Moezzi SMI, et al. (2021) Administration of COVID-19 vaccines in immunocompromised patients. Int Immunopharmacol 99: 108021. https://doi.org/10.1016/j.intimp.2021.108021
[151]	Zhao J, Zhao S, Ou J, et al. (2020) COVID-19: Coronavirus vaccine development updates. Front Immunol 11: 602256. https://doi.org/10.3389/fimmu.2020.602256
[152]	Chen Y, Shen H, Huang R, et al. (2021) Serum neutralising activity against SARS-CoV-2 variants elicited by CoronaVac. Lancet Infect Dis 21: 1071-1072. https://doi.org/10.1016/S1473-3099(21)00287-5
[153]	Olu-Abiodun O, Abiodun O, Okafor N, et al. (2022) COVID-19 vaccination in Nigeria: A rapid review of vaccine acceptance rate and the associated factors. PloS One 17: e0267691. https://doi.org/10.1371/journal.pone.0267691
[154]	Wu Z, Hu Y, Xu M, et al. (2021) Safety, tolerability, and immunogenicity of an inactivated SARS-CoV-2 vaccine (CoronaVac) in healthy adults aged 60 years and older: a randomised, double-blind, placebo-controlled, phase 1/2 clinical trial. Lancet Infect Dis 21: 803-812. https://doi.org/10.1016/s1473-3099(20)30987-7
[155]	Jafari A, Pouya FD, Niknam Z, et al. (2022) Current advances and challenges in COVID-19 vaccine development: from conventional vaccines to next-generation vaccine platforms. Mol Biol Rep 49: 4943-4957. https://doi.org/10.1007/s11033-022-07132-7
[156]	Trimpert J, Dietert K, Firsching TC, et al. (2021) Development of safe and highly protective live-attenuated SARS-CoV-2 vaccine candidates by genome recoding. Cell Rep 36: 109493. https://doi.org/10.1016/j.celrep.2021.109493
[157]	Jeyanathan M, Afkhami S, Smaill F, et al. (2020) Immunological considerations for COVID-19 vaccine strategies. Nat Rev Immunol 20: 615-632. https://doi.org/10.1038/s41577-020-00434-6
[158]	Li M, Wang H, Tian L, et al. (2022) COVID-19 vaccine development: milestones, lessons and prospects. Signal Transduct Target Ther 7: 146. https://doi.org/10.1038/s41392-022-00996-y
[159]	Folegatti PM, Ewer KJ, Aley PK, et al. (2020) Safety and immunogenicity of the ChAdOx1 nCoV-19 vaccine against SARS-CoV-2: a preliminary report of a phase 1/2, single-blind, randomised controlled trial. Lancet 396: 467-478. https://doi.org/10.1016/s0140-6736(20)31604-4
[160]	Zhang Z, Shen Q, Chang H, et al. (2022) Vaccines for COVID-19: A systematic review of immunogenicity, current development, and future prospects. Front Immunol 13: 843928. https://doi.org/10.3389/fimmu.2022.843928
[161]	Sah R, Shrestha S, Mehta R (2021) AZD1222 (Covishield) vaccination for COVID-19: Experiences, challenges, and solutions in Nepal. Travel Med Infect Dis 40: 101989. https://doi.org/10.1016/j.tmaid.2021.101989
[162]	Naskalska A, Dabrowska A, Nowak P, et al. (2018) Novel coronavirus-like particles targeting cells lining the respiratory tract. PloS One 13: e0203489. https://doi.org/10.1371/journal.pone.0203489
[163]	Ward BJ, Gobeil P, Séguin A, et al. (2021) Phase 1 randomized trial of a plant-derived virus-like particle vaccine for COVID-19. Nat Med 27: 1071-1078. https://doi.org/10.1038/s41591-021-01370-1
[164]	Fang E, Liu X, Li M, et al. (2022) Advances in COVID-19 mRNA vaccinedevelopment. Signal Transduct Target Ther 7: 94. https://doi.org/10.1038/s41392-022-00950-y
[165]	Collier ARY, Miller J, Hachmann NP, et al. (2023) Immunogenicity of the BA.5 bivalent mRNA vaccine boosters. New Engl J Med 388: 565-567. https://doi.org/10.1056/NEJMc2213948
[166]	Vogel AB, Kanevsky I, Che Y, et al. (2021) BNT162b vaccines protect rhesus macaques from SARS-CoV-2. Nature 592: 283-289. https://doi.org/10.1038/s41586-021-03275-y
[167]	Li M, Li Y, Li S, et al. (2022) The nano delivery systems and applications of mRNA. Eur J Med Chem 227: 113910. https://doi.org/10.1016/j.ejmech.2021.113910
[168]	Gooch KE, Smith TRF, Salguero FJ, et al. (2021) One or two dose regimen of the SARS-CoV-2 synthetic DNA vaccine INO-4800 protects against respiratory tract disease burden in nonhuman primate challenge model. Vaccine 39: 4885-4894. https://doi.org/10.1016/j.vaccine.2021.06.057
[169]	Kaur SP, Gupta V (2020) COVID-19 vaccine: a comprehensive status report. Virus Res 288: 198114. https://doi.org/10.1016/j.virusres.2020.198114
[170]	Tebas P, Yang S, Boyer JD, et al. (2021) Safety and immunogenicity of INO-4800 DNAvaccine against SARS-CoV-2: A preliminary report of an open-label, Phase 1 clinical trial. EClinical Medicine 31: 100689. https://doi.org/10.1016/j.eclinm.2020.100689
[171]	Brocato RL, Kwilas SA, Kim RK, et al. (2021) Protective efficacy of a SARS-CoV-2 DNA vaccine in wild-type and immunosuppressed Syrian hamsters. NPJ Vaccines 6: 16. https://doi.org/10.1038/s41541-020-00279-z
[172]	Mallory RM, Formica N, Pfeiffer S, et al. (2022) Safety and immunogenicity following a homologous booster dose of a SARS-CoV-2 recombinant spike protein vaccine (NVX-CoV2373): a secondary analysis of a randomised, placebo-controlled, phase 2 trial. Lancet Infect Dis 22: 1565-1576. https://doi.org/10.1016/s1473-3099(22)00420-0
[173]	Krammer F (2020) SARS-CoV-2 vaccines in development. Nature 586: 516-527. https://doi.org/10.1038/s41586-020-2798-3
[174]	Sinclair AH, Taylor MK, Weitz JS, et al. (2023) Reasons for receiving or not receiving bivalent COVID-19 booster vaccinations among adults—United States, November 1–December 10, 2022. MMWR Morb Mortal Wkly Rep 72: 73-75. https://doi.org/10.15585/mmwr.mm7203a5

This article has been cited by:

1.	V. Deeban Chakravarthy, K L. N. C. Prakash, Kadiyala Ramana, Thippa Reddy Gadekallu, 2023, Chapter 64, 978-981-19-3678-4, 733, 10.1007/978-981-19-3679-1_64
2.	Hao Zhang, Lina Ge, Guifen Zhang, Jingwei Fan, Denghui Li, Chenyang Xu, A two-stage intrusion detection method based on light gradient boosting machine and autoencoder, 2023, 20, 1551-0018, 6966, 10.3934/mbe.2023301
3.	Kadiyala Ramana, A. Revathi, A. Gayathri, Rutvij H. Jhaveri, C.V. Lakshmi Narayana, B. Naveen Kumar, WOGRU-IDS — An intelligent intrusion detection system for IoT assisted Wireless Sensor Networks, 2022, 196, 01403664, 195, 10.1016/j.comcom.2022.10.001
4.	Kranthi Kumar Singamaneni, Kadiyala Ramana, Gaurav Dhiman, Saurabh Singh, Byungun Yoon, A Novel Blockchain and Bi-Linear Polynomial-Based QCP-ABE Framework for Privacy and Security over the Complex Cloud Data, 2021, 21, 1424-8220, 7300, 10.3390/s21217300
5.	Olena Shlyakhetko, Michal Greguš, 2024, Chapter 5, 978-3-031-60814-8, 101, 10.1007/978-3-031-60815-5_5
6.	G. M. Nandana, Ashok Kumar Yadav, 2023, Chapter 8, 978-981-99-3314-3, 83, 10.1007/978-981-99-3315-0_8

Reader Comments

Your name:*

Email:*
© 2023 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)

通讯作者: 陈斌, bchen63@163.com

1.
沈阳化工大学材料科学与工程学院沈阳 110142

AIMS Allergy and Immunology

0.4

Metrics

Article views(2172) PDF downloads(200) Cited by(0)

Preview PDF

Download XML

Export Citation

Article outline

Show full outline

Figures and Tables

Figures(1) / Tables(1)

AIMS Allergy and Immunology

The journey so far with SARS-CoV-2 variants: Pathogenesis, immunity and treatments

Related Papers:

Abstract

1. Introduction

1.1. Signature-based/misuse detection mechanism

1.2. Anomaly-based detection mechanism

1.3. Hybrid detection mechanism

1.4. Limitations of current IDS

1.5. DataStream mining

2. Related works

3. Methodology

3.1. Data Preprocessing

3.2. One-hot encoding technique

3.3. Feature selection

3.4. Proposed classification algorithm (SAE-HT)

4. Experimental setup

4.1. Intrusion dataset

4.2. Performance evaluation

5. Results and discussion

5.1. Feature selection techniques analysis

5.2. Detection time analysis

5.3. Performance analysis of proposed classifier

5.4. Comparison with state-of-art techniques

6. Conclusions

Acknowledgments

Conflict of interest

Funding

References

This article has been cited by:

Reader Comments

通讯作者: 陈斌, bchen63@163.com

Metrics

Figures and Tables

Other Articles By Authors

Catalog

AIMS Allergy and Immunology

The journey so far with SARS-CoV-2 variants: Pathogenesis, immunity and treatments

Related Papers:

Abstract

1. Introduction

1.1. Signature-based/misuse detection mechanism

1.2. Anomaly-based detection mechanism

1.3. Hybrid detection mechanism

1.4. Limitations of current IDS

1.5. DataStream mining

2. Related works

3. Methodology

3.1. Data Preprocessing

3.2. One-hot encoding technique

3.3. Feature selection

3.4. Proposed classification algorithm (SAE-HT)

4. Experimental setup

4.1. Intrusion dataset

4.2. Performance evaluation

5. Results and discussion

5.1. Feature selection techniques analysis

5.2. Detection time analysis

5.3. Performance analysis of proposed classifier

5.4. Comparison with state-of-art techniques

6. Conclusions

Acknowledgments

Conflict of interest

Funding

References

This article has been cited by:

Reader Comments

通讯作者: 陈斌, bchen63@163.com

Metrics

Figures and Tables

Other Articles By Authors

Related pages

Tools

Export File

Citation

Format

Content

Catalog