Validation of CTL Epitope Prediction Methods
Author Information
Author(s): Mette V. Larsen, Claus Lundegaard, Kasper Lamberth, Soren Buus, Ole Lund, Morten Nielsen
Primary Institution: Technical University of Denmark
Hypothesis
How does the predictive performance of NetCTL-1.2 compare to other CTL epitope prediction methods?
Conclusion
NetCTL-1.2 shows higher predictive performance than EpiJen, MAPPP, MHC-pathway, and WAPP, although some comparisons are not statistically significant.
Supporting Evidence
- NetCTL-1.2 has a higher predictive performance than EpiJen, MAPPP, MHC-pathway, and WAPP on all performance measures.
- The higher performance of NetCTL-1.2 compared to EpiJen and MHC-pathway is not statistically significant on all measures.
- NetCTL-1.2 achieved a sensitivity among the 5% top-scoring peptides above 0.72.
Takeaway
This study tested different methods to predict immune responses to viruses, finding that one method, NetCTL-1.2, works better than others.
Methodology
The study compared the performance of five CTL epitope prediction methods using two evaluation datasets of known HIV CTL epitopes.
Limitations
The performance measures may not be applicable to all methods due to differences in output formats.
Statistical Information
P-Value
0.02
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website