CancerVaccine

CancerVaccine provides a curated library and predictive machine learning models to support design, evaluation, and prioritization of peptide-based cancer vaccines.

Key Features:

• Library Construction: A database of peptide-based cancer vaccines annotated with clinical performance and related clinical attributes.
• Classification Based on Clinical Response: Vaccines are categorized as high clinical response (HCR) or low clinical response (LCR) based on observed efficacy in clinical settings.
• Focus on Modified Peptides: Includes analysis of modified peptides derived from artificially altered proteins with reported promise for melanoma vaccines.
• HLA Class II Affinity Screening: Screening for HLA class II affinity peptides to match peptides with appropriate HLA alleles for T cell responses.
• Adjuvant Potential: Identification of adjuvant associations, including Montanide ISA-51, that correlate with improved clinical responses.
• Machine Learning Model: A high sensitivity and specificity machine learning model that integrates peptide type, matched HLA allele, adjuvant choice, and treatment regimen to predict clinical response.

Scientific Applications:

• Vaccine Candidate Prioritization: Prioritizes peptide vaccine candidates for further experimental evaluation based on clinical-response classification and predictive scores.
• T Cell Epitope Design: Informs design of T cell epitope peptides by linking peptide properties and HLA class II affinity to clinical outcomes.
• Melanoma Vaccine Development: Supports selection and study of modified peptides with reported relevance for melanoma immunotherapy.
• Adjuvant and Regimen Selection: Guides selection of adjuvants (e.g., Montanide ISA-51) and treatment regimens associated with higher clinical responses.
• Experimental Design and Translation: Provides structured data and predictive models to generate hypotheses and guide translational cancer immunotherapy studies.

Methodology:

Constructed a database of peptide-based cancer vaccines annotated with clinical performance; classified entries into high clinical response (HCR) and low clinical response (LCR); performed HLA class II affinity screening of peptides; analyzed modified peptides; and developed a machine learning model (reported high sensitivity and specificity) integrating peptide type, matched HLA allele, adjuvant (e.g., Montanide ISA-51), and treatment regimen.