PEP-PRED Na+

PEP-PRED Na+ predicts peptides that block voltage-gated Na+ channels to identify candidate therapeutic blockers for diseases associated with Na+ channel dysfunction.

Key Features:

• Target specificity: Predicts peptides that specifically target and block voltage-gated sodium (Na+) channels.
• Machine learning model: Implements a random forest ensemble classifier for peptide classification.
• Cross-validation performance: Cross-validation results: Sensitivity = 0.81, Accuracy = 0.83, Precision = 0.85, F-score = 0.83, Specificity = 0.86, Matthew's correlation coefficient = 0.67.
• Testing performance: Independent testing results: Sensitivity = 0.88, Accuracy = 0.92, Precision = 0.96, F-score = 0.91, Specificity = 0.96, Matthew's correlation coefficient = 0.84.
• Peptide sources: Considers peptides derived from diverse biological sources, including venoms.

Scientific Applications:

• Epilepsy: Provide candidate Na+ channel blocking peptides for therapeutic development against epilepsy.
• Chronic pain: Identify peptide candidates that may modulate pain via Na+ channel blockade.
• Cardiovascular disorders: Supply potential peptide blockers for Na+ channel–related cardiovascular conditions.
• Cancer: Deliver candidate peptides for investigating Na+ channel involvement in cancers.
• Immune system abnormalities: Offer peptide candidates to study Na+ channel roles in immune dysfunction.
• Neuromuscular issues: Generate candidate blockers for neuromuscular disorders involving Na+ channels.
• Respiratory conditions: Produce peptide candidates to explore Na+ channel–linked respiratory pathologies.

Methodology:

Uses a random forest classifier trained and evaluated via cross-validation and independent testing, with reported sensitivity, accuracy, precision, F-score, specificity, and Matthew's correlation coefficient.