PPI-Miner: A Structure and Sequence Motif Co-Driven Protein-Protein Interaction Mining and Modeling Computational...

Area of Science:

• Computational biology
• Structural biology
• Bioinformatics

Background:

• Protein-protein interactions (PPIs) are crucial for biological processes.
• Identifying PPIs via biochemical methods is challenging due to the involvement of small protein motifs.
• Existing methods often require extensive cloning of mutant proteins.

Purpose of the Study:

• To present PPI-Miner, a novel computational method for predicting protein-protein interactions using protein motifs.
• To develop a motif-matching algorithm for identifying proteins with similar motifs.
• To apply PPI-Miner for discovering potential substrates of cereblon (CRBN).

Main Methods:

• Developed a motif-matching algorithm to find proteins with sequential or structural motif similarity.
• Determined binding modes of motifs and receptor proteins to construct PPI complexes.
• Utilized an automated protocol for building and optimizing PPI complex structures.
• Applied PPI-Miner to the human proteome to predict CRBN substrates.

Main Results:

• PPI-Miner successfully identified 1,739 potential cereblon (CRBN) substrates from the human proteome.
• 16 of the predicted CRBN substrates were experimentally validated in prior studies.
• The method demonstrates potential for applications beyond PPI prediction, including molecular glues and vaccine design.

Conclusions:

• PPI-Miner offers an efficient computational approach for predicting protein-protein interactions based on protein motifs.
• The tool facilitates the discovery of novel protein interactions and substrate identification, as shown by the CRBN substrate prediction.
• PPI-Miner provides a valuable resource for biological research and drug discovery, with accessible code and web services.