iProt-Sub

iProt-Sub predicts protease-specific substrates and their cleavage sites across Aspartic (A), Cysteine (C), Metallo (M), and Serine (S) protease families for proteome-wide analysis.

Key Features:

• Protease coverage: Predicts cleavage sites for 38 distinct proteases spanning Aspartic (A), Cysteine (C), Metallo (M), and Serine (S) families.
• Optimized prediction models: Uses optimized cleavage-site prediction models trained on an extensive dataset of protease-specific cleavage events to improve performance and specificity.
• Sequence and structural feature integration: Integrates diverse sequence and structural features to represent substrate-protease interactions.
• Eleven sequence encoding schemes: Employs 11 sequence encoding schemes including local amino acid sequence profiles, secondary structure predictions, solvent accessibility, and native disorder information.
• Two-step feature selection: Applies a two-step feature selection procedure to remove redundant and irrelevant features and retain informative predictors.
• Benchmarking and validation: Performance assessed by cross-validation and independent tests and compared against existing generic tools.

Scientific Applications:

• Proteome-wide substrate prediction: Enables large-scale prediction of protease cleavage sites across proteomes.
• Substrate specificity analysis: Facilitates hypothesis-driven investigations into protease substrate specificity and cleavage preferences.
• Proteolytic regulation studies: Supports studies of proteolytic events and regulatory mechanisms underlying cellular processes influenced by proteolysis.

Methodology:

Optimized cleavage-site prediction models trained on data for ~38 proteases; integration of sequence and structural features using 11 encoding schemes (local amino acid sequence profiles, secondary structure predictions, solvent accessibility, native disorder); two-step feature selection; benchmarking via cross-validation and independent tests.