Co-Mutations and Possible Variation Tendency of the Spike RBD and Membrane Protein in SARS-CoV-2 by Machine Learning

Area of Science:

• Virology
• Molecular Biology
• Genomics

Background:

• Emergence of SARS-CoV-2 variants with mutations in spike RBD and M protein.
• These mutations may enhance immune evasion and breakthrough infections.

Purpose of the Study:

• Investigate co-mutations between spike RBD and M protein.
• Elucidate the relationship and potential synergy between these mutations.

Main Methods:

• Utilized EVmutation for mutation distribution analysis.
• Employed Sequence-to-Sequence Transformer Model (S2STM) for sequence mapping.
• Analyzed specific spike RBD mutation sets (G339D-S373P-S375F, Q493R-Q498R-Y505).

Main Results:

• Identified specific spike RBD mutations correlating with M protein alterations at sites 3 and 19/63.
• Demonstrated a propensity for certain spike RBD mutations to induce M protein changes.
• Established a mapping between spike RBD and M protein sequences.

Conclusions:

• Highlighted mutational synergy between SARS-CoV-2 spike RBD and M proteins.
• Proposed a potential mechanism for viral evolution driven by these co-mutations.
• Provided a novel approach for analyzing interrelationships between viral protein sequences.