VarEPS: an evaluation and prewarning system of known and virtual variations of SARS-CoV-2 genomes

Qinglan Sun, Chang Shu, Wenyu Shi, Yingfeng Luo, Guomei Fan, Jingyi Nie, Yuhai Bi, Qihui Wang, Jianxun Qi, Jian Lu, Yuanchun Zhou, Zhihong Shen, Zhen Meng, Xinjiao Zhang, Zhengfei Yu, Shenghan Gao, Linhuan Wu, Juncai Ma, Songnian HuNucleic Acids Res 2022 · DOI · OpenReview

TL;DR

The COVID-19 pandemic is driven by the rapid evolution of SARS-CoV-2. To combat this, researchers have developed VarEPS, a comprehensive system that evaluates the potential impact of both known and hypothetical viral mutations on transmissibility, antibody binding, and diagnostic primer efficacy. This tool aims to proactively identify dangerous variants and guide the development of future vaccines and therapeutics.

What Problem Does This Paper Address?

The SARS-CoV-2 virus is an RNA virus, which means it mutates at a relatively high rate. These mutations can lead to new variants with altered characteristics, such as increased transmissibility, virulence, or the ability to evade immune responses generated by vaccines or prior infection. Monitoring these mutations is crucial for understanding the pandemic's trajectory and for developing effective countermeasures. While existing methods often rely on tracking mutation frequency, this paper addresses the need for a more sophisticated approach that predicts the functional impact of these mutations, including those that haven't yet emerged (virtual mutations). This predictive capability is vital for staying ahead of the virus and for designing robust vaccines and treatments that are less susceptible to viral escape.

Methods

VarEPS employs a multi-faceted approach to evaluate viral variants:

  • Genomic and Structural Analysis: The system analyzes known and virtual mutations for their effects on key viral properties, including:
  • Physicochemical properties
    • Translation efficiency
    • Secondary structure
    • Binding capacity to ACE2 (the virus's entry point into cells)
    • Binding capacity to neutralizing antibodies
  • AI-Based Risk Prediction: A machine learning algorithm, specifically a random forest classifier, is used to verify the effectiveness of these genomic and structural characteristics for predicting risk. This classifier can group viral strains based on their predicted transmissibility and affinity to neutralizing antibodies.
  • Primer Efficacy Evaluation: The system assesses how mutations might affect the performance of RT-PCR diagnostic primers, warning of potential false negatives.
  • Virtual Mutation Analysis: A key innovation is the inclusion of "virtual" mutations, allowing for the evaluation of potential future variants before they become widespread.

Datasets

The study utilized the following datasets:

  • SARS-CoV-2 Genome Sequences:
    • Approximately 2.6 million sequences from the GISAID database.
  • Approximately 0.96 million sequences from the US National Center for Biotechnology Information (NCBI) and the National Microbiology Data Center (NMDC).
  • Primer Information: Data from the USA Centers for Disease Control and Prevention (CDC), the Chinese Center for Disease Control and Prevention (CDC China), and the World Health Organization (WHO).

The paper does not introduce new benchmark datasets but rather leverages existing large-scale genomic data for its analysis and system development.

Key Insights and Ideas

  • Beyond Frequency: The most significant insight is the shift from solely relying on mutation frequency to a functional, impact-based evaluation. This allows for the proactive identification of potentially dangerous variants, even if they are currently rare or haven't emerged yet.
  • Predicting Immune Evasion: The system's ability to predict how mutations affect antibody binding is crucial for understanding potential vaccine escape and for designing antibodies that are broadly effective against future variants.
  • Virtual Variants as Early Warnings: Evaluating "virtual" mutations provides a powerful foresight mechanism, allowing researchers to anticipate and prepare for potential future threats posed by novel viral strains.
  • Integrated Risk Assessment: VarEPS consolidates multiple lines of evidence (genomic, structural, AI-based) to provide a holistic risk assessment for viral variants, which is more robust than single-metric approaches.

Key Results

  • VarEPS was established, incorporating both known and virtual mutations of SARS-CoV-2 genomes.
  • The system comprehensively analyzes the effects of mutations on physicochemical properties, translation efficiency, secondary structure, ACE2 binding, and neutralizing antibody binding.
  • An AI-based algorithm was developed and used to group viral strains by transmissibility and affinity to neutralizing antibodies.
  • The system provides a unique resource for quickly evaluating the variation risks of key sites, aiding in vaccine and drug development.
  • The database is freely accessible at www.nmdc.cn/ncovn.

Limitations and Future Work

  • In Silico vs. In Vivo: The paper acknowledges that the current models are based on in silico predictions. Future work will involve iterative optimization of these models through experimental validation (in vitro and in vivo studies) to improve the accuracy of simulations and predictions.
  • Continuous Updates: The authors plan to continuously update the system with new data on SARS-CoV-2 genome sequences to maintain its relevance and accuracy.
  • Broader Evaluations: The machine learning model, which is presented as the first to successfully evaluate binding affinity and group strains, will be further developed for broader evaluations.
  • Diagnostic Primer Impact: While the system flags potential issues with primer efficacy, the authors note that many molecular tests can tolerate minor mismatches, and the practical impact of these mismatches needs to be considered in context.

Summary updated

Abstract

The genomic variations of SARS-CoV-2 continue to emerge and spread worldwide. Some mutant strains show increased transmissibility and virulence, which may cause reduced protection provided by vaccines. Thus, it is necessary to continuously monitor and analyze the genomic variations of SARS-COV-2 genomes. We established an evaluation and prewarning system, SARS-CoV-2 variations evaluation and prewarning system (VarEPS), including known and virtual mutations of SARS-CoV-2 genomes to achieve rapid evaluation of the risks posed by mutant strains. From the perspective of genomics and structural biology, the database comprehensively analyzes the effects of known variations and virtual variations on physicochemical properties, translation efficiency, secondary structure, and binding capacity of ACE2 and neutralizing antibodies. An AI-based algorithm was used to verify the effectiveness of these genomics and structural biology characteristic quantities for risk prediction. This classifier could be further used to group viral strains by their transmissibility and affinity to neutralizing antibodies. This unique resource makes it possible to quickly evaluate the variation risks of key sites, and guide the research and development of vaccines and drugs. The database is freely accessible at www.nmdc.cn/ncovn.
BibTeX
@article{sun2022vareps,
  author = {Qinglan Sun and Chang Shu and Wenyu Shi and Yingfeng Luo and Guomei Fan and Jingyi Nie and Yuhai Bi and Qihui Wang and Jianxun Qi and Jian Lu and Yuanchun Zhou and Zhihong Shen and Zhen Meng and Xinjiao Zhang and Zhengfei Yu and Shenghan Gao and Linhuan Wu and Juncai Ma and Songnian Hu},
  title = {{VarEPS: an evaluation and prewarning system of known and virtual variations of SARS-CoV-2 genomes}},
  journal = {Nucleic Acids Res.},
  year = {2022},
  doi = {10.1093/nar/gkab921},
}
Figure