Comprehensive classification of proteins based on structures that engage lipids by COMPOSEL

Area of Science:

• Structural biology
• Computational biology
• Biochemistry

Background:

• Protein structure prediction, including for membrane proteins, relies on AI/ML models trained on experimental data.
• Accurate prediction requires incorporating restraints, especially for membrane proteins whose function is tied to lipid bilayers.
• Existing classifications for membrane proteins do not fully capture their lipid interactions.

Purpose of the Study:

• To introduce a novel classification system, COMPOSEL (Classification Of Membrane Proteins based On Structures Engaging Lipids), for membrane proteins.
• To enhance AI/ML-driven structure prediction by integrating lipid environment and protein architecture details.
• To provide a framework for describing protein function, signaling, and interactions within the cellular lipid context.

Main Methods:

• Development of the COMPOSEL nomenclature, building upon existing membrane protein classifications (monotopic, bitopic, polytopic, peripheral) and lipid types.
• Integration of functional and regulatory elements, including lipid recognition sites and binding partners.
• Demonstration of COMPOSEL's application using diverse examples like synaptotagmins, PDZD8, Protrudin, MARCKS, caveolins, BAM, aGPCRs, DGKε, and FALDH.

Main Results:

• COMPOSEL effectively describes lipid interactivity, signaling mechanisms, and binding of various molecules (metabolites, drugs, nucleic acids) for different proteins.
• The system demonstrates scalability to represent genomic encoding of membrane structures and pathogen interactions.
• Examples showcase COMPOSEL's ability to detail specific protein-lipid engagements and functional consequences.

Conclusions:

• COMPOSEL offers a comprehensive approach to classifying membrane proteins based on their structural engagement with lipids.
• This classification system can significantly improve the accuracy and scope of AI/ML-based protein structure prediction.
• COMPOSEL provides a unified framework for understanding membrane protein function, cellular processes, and even disease mechanisms.