Evolutionary artificial intelligence based peptide discoveries for effective Covid-19 therapeutics

Area of Science:

• Computational biology
• Machine learning
• Virology
• Drug discovery

Background:

• The COVID-19 pandemic necessitated rapid development of therapeutic strategies.
• Computational biology and machine learning offer accelerated drug discovery approaches.

Purpose of the Study:

• To identify conserved motifs in diverse SARS-CoV-2 sequences using machine learning.
• To design and evaluate peptides targeting the main protease (Mpro) of SARS-CoV-2.

Main Methods:

• Machine learning algorithms applied to viral sequences from India, China, Italy, and USA.
• Identification of conserved motifs and design of a peptide library.
• Molecular Dynamics (MD) simulations to assess peptide binding affinity, stability, and specificity.

Main Results:

• Conserved motifs were identified across diverse COVID-19 viral strains.
• Four designed peptides demonstrated strong binding affinity to the SARS-CoV-2 main protease (Mpro).
• MD simulations confirmed the stability and specificity of these peptides under physiological conditions.

Conclusions:

• The identified evolutionary peptides show promise as potential therapeutics against COVID-19.
• These peptides may offer cross-protection against various SARS-CoV-2 variants.
• This study highlights the utility of machine learning in rapid therapeutic development for emerging infectious diseases.