Computational generation of an annotated gigalibrary of synthesizable, composite peptidic macrocycles

Peptidomimetic macrocycles have the potential to regulate challenging therapeutic targets. Structures of this type having precise shapes and drug-like character are particularly coveted, but are relatively difficult to synthesize. Our laboratory has developed robust methods that integrate small-pept...

Full description

Saved in:
Bibliographic Details
Published in:Proceedings of the National Academy of Sciences - PNAS 2020-10, Vol.117 (40), p.24679-24690
Main Authors: Saha, Ishika, Dang, Eric K., Svatunek, Dennis, Houk, Kendall N., Harran, Patrick G.
Format: Article
Language:English
Subjects:
Biological Sciences
Computer applications
Libraries
Methodology
Parallel processing
Peptides
Physical Sciences
Source code
Synthesis
Three dimensional composites
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Peptidomimetic macrocycles have the potential to regulate challenging therapeutic targets. Structures of this type having precise shapes and drug-like character are particularly coveted, but are relatively difficult to synthesize. Our laboratory has developed robust methods that integrate small-peptide units into designed scaffolds. These methods create macrocycles and embed condensed heterocycles to diversify outcomes and improve pharmacological properties. The hypothetical scope of the methodology is vast and far outpaces the capacity of our experimental format. We now describe a computational rendering of our methodology that creates an in silico three-dimensional library of composite peptidic macrocycles. Our open-source platform, CPMG (Composite Peptide Macrocycle Generator), has algorithmically generated a library of 2,020,794,198 macrocycles that can result from the multistep reaction sequences we have developed. Structures are generated based on predicted site reactivity and filtered on the basis of physical and three-dimensional properties to identify maximally diverse compounds for prioritization. For conformational analyses, we also introduce ConfBuster++, an RDKit port of the open-source software ConfBuster, which allows facile integration with CPMG and ready parallelization for better scalability. Our approach deeply probes ligand space accessible via our synthetic methodology and provides a resource for large-scale virtual screening.
ISSN:0027-8424
1091-6490
DOI:10.1073/pnas.2007304117