NetCTLpan

NetCTLpan predicts MHC class I-presented cytotoxic T lymphocyte (CTL) epitopes by integrating proteasomal cleavage, TAP transport efficiency, and MHC class I binding affinity into a pathway likelihood score.

Key Features:

• Integrated Predictive Model: Combines predictions of proteasomal cleavage, TAP transport efficiency, and MHC class I binding affinity into a comprehensive MHC class I pathway likelihood score.
• Pan-Specificity: Makes pan-specific predictions for all known MHC class I molecules based on their protein sequences and supports peptide lengths of 8–11 amino acids (mers).
• Optimization for Specificity: Prioritizes high specificity to minimize false positives through methodological optimization.
• Training and Validation: Was trained and validated on extensive datasets of experimentally identified MHC class I ligands and CTL epitopes, underpinning reported superior predictive performance versus other methods.
• Fixed Weights for Predictive Components: Uses fixed weights for proteasomal cleavage and TAP transport across all MHC molecules following analysis that found no consistent MHC-dependent weight patterns.

Scientific Applications:

• Rational Vaccine Design: Assists rational vaccine design by identifying candidate immunogenic peptides predicted to be processed and presented by MHC class I molecules.
• Reduction in Experimental Effort: Has been reported to reduce the experimental effort required to identify 90% of new epitopes by approximately 15% and 40%, compared to NetMHCpan and NetCTL.

Methodology:

Integrates proteasomal cleavage, TAP transport efficiency, and MHC class I binding affinity predictions into a combined MHC class I pathway likelihood score, employs fixed weights for proteasomal cleavage and TAP transport, and was trained and validated on datasets of experimentally identified MHC class I ligands and CTL epitopes.