Procleave: Predicting Protease-specific Substrate Cleavage Sites by Combining Sequence and Structural Information

Area of Science:

• Biochemistry and Bioinformatics
• Enzymology
• Computational Biology

Background:

• Proteases are crucial enzymes regulating protein degradation and physiological processes.
• Accurate identification of protease substrates is vital for understanding enzyme function and developing therapeutics.
• Existing bioinformatics methods often lack accuracy in predicting cleavage sites, necessitating improved approaches.

Purpose of the Study:

• To develop a novel bioinformatics approach, Procleave, for predicting protease-specific substrates and cleavage sites.
• To leverage both sequence and 3D structural information for enhanced prediction accuracy.
• To provide a publicly accessible webserver for researchers.

Main Methods:

• Developed Procleave, a bioinformatics tool integrating sequence and 3D structural data.
• Utilized LOWESS data-smoothing to represent structural features of cleavage sites.
• Employed a conditional random field (CRF) framework incorporating sequence, chemical group, and structural features.

Main Results:

• Procleave demonstrated outstanding performance in benchmarking and independent tests.
• The tool accurately identified most cleavage sites in a case study.
• Application to the human structural proteome revealed potential novel protease targets and cleavage sites.

Conclusions:

• Procleave offers a highly accurate method for predicting protease cleavage sites by integrating structural and sequence information.
• The tool has significant implications for understanding protease functions and identifying new therapeutic targets.
• Procleave is available as a webserver, facilitating broader research applications.