PepX: a structural database of non-redundant protein-peptide complexes

Area of Science:

• Structural Biology
• Biochemistry
• Drug Discovery

Background:

• Protein-peptide interactions are vital in cellular signaling and regulation, representing up to 40% of all protein interactions.
• Structural information for these interactions is limited, hindering their use in drug design.
• Existing datasets are often biased or restricted to small ligands, lacking comprehensive structural diversity.

Purpose of the Study:

• To establish an unbiased and exhaustive dataset of protein-peptide complexes.
• To provide a foundation for modeling and designing peptide-mediated therapeutics.
• To enable exploration of structurally diverse protein-peptide interactions.

Main Methods:

• Compiled all protein-peptide complexes from the Protein Data Bank (PDB) with peptide lengths up to 35 residues.
• Clustered complexes based on binding interfaces, not sequence homology, to ensure structural diversity.
• Annotated complexes with biological and structural information and incorporated backbone variations from the BriX database.

Main Results:

• Developed PepX, a dataset containing 1431 complexes.
• Generated 505 unique protein-peptide interface clusters, representing a wide range of structural interactions.
• Annotated peptides with naturally occurring backbone variations for enhanced peptide design.

Conclusions:

• PepX provides a comprehensive and unbiased resource for studying protein-peptide interactions.
• The dataset facilitates the identification of novel therapeutic targets and the design of peptide-based drugs.
• Structural diversity and detailed annotations in PepX are crucial for advancing the field of peptide drug discovery.