A hybrid computational framework for intelligent inter-continent SARS-CoV-2 sub-strains characterization and...

Area of Science:

• Genomics
• Virology
• Computational Biology

Background:

• Early SARS-CoV-2 (Severe Acute Respiratory Syndrome Coronavirus 2) spread lacked clarity on pathogenicity and sub-strain origins.
• Understanding viral evolution is crucial for predicting and managing future outbreaks.

Purpose of the Study:

• To analyze SARS-CoV-2 genome diversity and identify transmission patterns of sub-strains.
• To develop a robust system for predicting emerging viral variants using machine learning.

Main Methods:

• Analysis of 8864 human SARS-CoV-2 complete genome sequences from the GISAID database (December 2019 - January 2021).
• Application of a hybrid approach combining biotechnology and machine learning for genome diversity and pattern correlation analysis.
• Utilized a novel cognitive approach for knowledge mining to discover transmission routes.

Main Results:

• Corroborated the emergence and increase of inter- and intra- SARS-CoV-2 sub-strains across continents.
• Observed dynamic nucleotide mutations and transformation of viral patterns into new sub-strain clusters.
• Developed a superior classification system for predicting new viral sub-strains compared to state-of-the-art methods.

Conclusions:

• The study provides insights into SARS-CoV-2 sub-strain dynamics, explaining concerns about the virus and potential future waves.
• The findings support enhanced contact tracing and prediction of viral evolution.
• Future work aims to refine intra-country sub-strain analytics for greater precision.