Procleave

Procleave predicts protease substrate cleavage sites by integrating sequence, chemical group-based, and 3D structural features using a conditional random field framework.

Key Features:

• Integration of Sequence and Structural Data: Utilizes a conditional random field (CRF) framework to combine sequence data with 3D structural features of substrates to enhance cleavage-site prediction.
• Structural Feature Representation: Applies LOWESS data-smoothing to represent structural features of known cleavage sites as discrete values for model input.
• Extensive Benchmarking and Validation: Demonstrated high accuracy in identifying cleavage sites through rigorous benchmarking, independent testing, and case studies.

Scientific Applications:

• Understanding Protease Functions: Aids elucidation of physiological roles of proteases by predicting substrate cleavage sites involved in protein degradation pathways.
• Therapeutic Target Identification: Suggests potential novel target substrates and corresponding cleavage sites across different proteases relevant to drug discovery.
• Pharmaceutical Applicability: Informs development of protease inhibitors or modulators by providing predicted cleavage sites for targeted proteolytic processes.

Methodology:

Maps substrates from the MEROPS database onto the Protein Data Bank (PDB) to obtain protein substrates with solved 3D structures; encodes structural parameters of cleavage sites within a CRF alongside sequence and chemical group-based features; represents structural features as discrete values using LOWESS data-smoothing.