NetTepi 1.0 Server
NetTepi 1.0 predicts T-cell epitopes from protein sequences. The method integrates
three prediction types, peptide-MHC binding affinity, peptide-MHC stability and
The server allows for predictions of T-cell epitopes restricted
to 13 different Human MHC (HLA) alleles representing 11 of the 12 common HLA-A and
B Supertypes as defined by Lund et al (2004).
Predictions of lengths 8-14: Predictions can be made for lengths between
8 and 14 for all alleles using an approximation algorithm. Note that only lengths
9 and 10 have been thoroughly benchmarked. Caution should be taken with predictions
for all other lengths.
Prediction values are calculated as a weighted sum of binding affinity, stability
and T-cell propensity prediction scores. A % Rank score based on predictions for
200.000 random natural peptides is also provided.
Peptide-MHC binding affinity predictions are obtained using the
Peptide-MHC stability predictions are obtained using the
T-cell propensity is predicted using the immunogenicity model described by
Calis et al (2013).
Note as of 5 dec 2016 the server has been updated to make affinity
predictions using NetMHCcons-1.1. Old functionality can be recovered following
the link Earlier functionality.
For publication of results, please cite:
NetTepi: an integrated method for the prediction of T cell epitopes.
Trolle T, Nielsen M. Immunogenetics (2014). 66(7-8):449-56.