Computational cyclic peptide design machine learning & Rosetta based methods.

Area of Science:

• Biochemistry
• Medicinal Chemistry
• Drug Discovery

Background:

• Macrocyclic peptides present a viable alternative to traditional therapeutics like antibodies and small molecules.
• They offer enhanced stability and binding specificity, particularly for targeting challenging intracellular surfaces.
• Head-to-tail cyclic peptides provide increased proteolytic protection and allow straightforward incorporation of non-canonical amino acids.

Purpose of the Study:

• To review current computational and experimental algorithms for the rational design of cyclic peptides.
• To highlight the challenges associated with cyclic peptide design, including conformational constraints and backbone sampling.
• To provide insights into advancing the development of macrocyclic peptide therapeutics.

Main Methods:

• Review of existing literature on computational algorithms for cyclic peptide design.
• Analysis of experimental techniques employed in cyclic peptide synthesis and characterization.
• Examination of methods for conformational analysis and structure-based design.

Main Results:

• Macrocyclic peptides demonstrate significant potential for targeting previously undruggable biological targets.
• Chemical synthesis facilitates the incorporation of diverse amino acids, expanding design possibilities.
• Despite design challenges, advancements in computational and experimental approaches are enabling more effective cyclic peptide development.

Conclusions:

• Macrocyclic peptides are a promising class of therapeutics with advantages over traditional drugs.
• Overcoming design complexities is key to realizing their full therapeutic potential.
• Continued development of design algorithms will accelerate the application of macrocyclic peptides in medicine.