Machine Learning to Identify Critical Biomarker Profiles in New SARS-CoV-2 Variants

Area of Science:

• Molecular Biology
• Virology
• Genomics

Background:

• The emergence of SARS-CoV-2 variants (Alpha, Beta, Omicron) poses global health challenges.
• Understanding host-pathogen interactions is crucial for combating viral infections.
• Eukaryotic translation factors play fundamental roles in protein synthesis and cellular processes.

Purpose of the Study:

• To investigate the relationship between SARS-CoV-2 variants and host eukaryotic translation factors.
• To determine if translation factor expression profiles can differentiate between variants and vaccination statuses.
• To identify causal relationships among translation factors in the context of SARS-CoV-2 infection.

Main Methods:

• Statistical analysis of eukaryotic translation factor expression profiles in infected individuals.
• Comparative analysis across different SARS-CoV-2 variants (Alpha, Beta, Omicron).
• Assessment of differences between vaccinated and non-vaccinated infected individuals.

Main Results:

• Successfully differentiated between SARS-CoV-2 variants based on translation factor expression.
• Partially differentiated between vaccinated and non-vaccinated infected individuals using the same profiles.
• Identified common causal relationships among translation factors, highlighting their interplay with the virus.

Conclusions:

• Host translation factor expression profiles serve as potential biomarkers for SARS-CoV-2 variant identification.
• Vaccination status can be inferred, to some extent, from host translation factor expression.
• The study elucidates the intricate interplay between SARS-CoV-2 variants and the host's protein synthesis machinery.