This record provides bibliographic details and links to the original research. An editorial study breakdown is not available.
What the researchers found
Machine learning and Rosetta-based computational methods can now design cyclic peptides that target flat protein surfaces considered undruggable by small molecules.
Why it matters
Cyclic peptides bridge the gap between small drugs and antibodies. Better design tools could accelerate development of drugs for currently untreatable diseases.
The numbers in context
Macrocyclic peptides target flat/intracellular undruggable surfaces; enhanced proteolytic stability; non-canonical amino acids incorporated; Rosetta and ML-based design methods reviewed
How the study worked
Review of computational algorithms including Rosetta-based methods, machine learning approaches, and experimental validation strategies for cyclic peptide design.
Who was studied
N/A (review of computational peptide design methods)
What this study cannot tell us
Methodology review focused on computational approaches. Many designed peptides still need experimental validation. Computational predictions do not always match in vitro activity.
Read the original research
Computational cyclic peptide design machine learning & Rosetta based methods.
Methods in enzymology, 723, 455-476
Citation
Sarmeili, Faraz; Siegler, Hannah; Powers, Andrew C; Hosseinzadeh, Parisa. (2025). Computational cyclic peptide design machine learning & Rosetta based methods.. Methods in enzymology, 723, 455-476. https://doi.org/10.1016/bs.mie.2025.09.016