|
|
|
The database is based on previous publications on T-cell epitopes and MHC ligands.
Howto cite the database:
Hans-Georg Rammensee, Jutta Bachmann, Niels Nikolaus Emmerich, Oskar Alexander Bachor, Stefan Stevanovic: SYFPEITHI: database for MHC ligands and peptide motifs. Immunogenetics (1999) 50: 213-219 (access via : www.syfpeithi.de)
An X in peptide sequences stands for an undetermined amino acid. However, if the peptide sequence has been determined by mass spectometry, as is the case for the peptides reported by Hunt and co-workers, X stands for either Leu or Ile (which have the same mass).
As far as T-cell epitopes are concerned, only those have been selected which are likely to be naturally processed.
From the numerous class II
motifs that have been published, we selected the more convincing ones,
i.e. those compatible with class II structure. Due to the variable number
of amino acids between the N-terminus and the first anchor of peptides,
alignment of ligands or T-cell epitopes to class II motifs is always arbitrary,
unless a structural analysis has been performed. In case of class II ligands
occurring as nested sets, we included only one or a few members of the
set in many cases.
Specific information on MHC motifs included in the database
References without link to Pubmed:
Brander and Walker 1996: Brander, C. and Walker, B. The HLA-class I CTL response in HIV-1 infection: Identification of optimal epitopes, Los Alamos, New Mexico: Los Alamos National Laboratory, Theoretical Biology and Biophysics 1996
Stevanovic et al. 1995: Stevanovic, S., Pomer, S., and Rammensee, H.-G. Oberflächenantigene im Nierenzellkarzinom - Präsentation von MHC I-gebundenen Selbstpeptiden. Akt. Urol. Sonderheft (26): 45-46, 1995
Details on motifs included in the database:
RT1.Aa (TAP2A-associated): Residues were determined by acid elution and pool sequencing of bound peptides following in vitro refolding of the recombinant alpha chain around a random peptide library.
RT1.Aa (TAP2B-associated): Residues were determined by acid elution and pool sequencing of bound peptides following in vitro refolding of the recombinant alpha chain around a random peptide library.
Mafa-A2: T-cell epitope determined by serological methods
The scoring system evaluates every amino acid within a given peptide. Individual amino acids may be given the arbitrary value 1 for amino acids that are only slightly preferred in the respective position, optimal anchor residues are given the value 15; any value between these two is possible. Negative values are also possible for amino acids which are disadvantageous for the peptide's binding capacity at a certain sequence position. The allocation of values is based on the frequency of the respective amino acid in natural ligands, T-cell epitopes, or binding peptides. Further explanations on the algorithm can be found in HG Rammensee, J Bachmann, NPN Emmerich, OA Bachor and S Stevanovic (1999) SYFPEITHI: database for MHC ligands and peptide motifs. Immunogenetics 50: 213-219
Our main aim is to have an extremely reliable SYFPEITHI epitope prediction algorithm. Therefore, only those MHC class I alleles for which a large amount of data is available are included in the "epitope prediction" section of SYFPEITHI. A reliability of at least 80% in retrieving the most apt epitope can be expected. Thus the naturally presented epitope should be among the top-scoring 2 % of all peptides predicted in 80 % of all predictions. For example, if a protein sequence is as long as 400 amino acids, it can be split into 392 possible nonameric peptides. Therefore, the correct epitope should then be among the 8 top-scoring peptides.
For epitope predictions using MHC class II motifs high reliabilities usually cannot be achieved because of the more variable pocket binding behavior. We estimate a reliability of approximately 50% only.
Please note:
· Different MHC class
I molecules prefer a different length of ligands. For example, SYFPEITHI
offers predictions for H2-Kb octamers, HLA-A*0201 nonamers and decamers,
or HLA-B8 octamers and nonamers.
· The maximal scores
vary between different MHC alleles. Therefore, we advise you to include
known ligands/epitopes in order to have an approximation of the scoring.
For example, the maximal score for HLA-A*0201 peptides is 36. The well-known
epitope GILGFVFTL derived from the influenza A matrix protein scores 30.
· All predicted MHC
class II ligands are 15mers, consisting of three N-terminal flanking residues,
the nonamer core sequence located within the binding groove, and three
C-terminal flanking residues. Thus, anchor residue P1 appears in position
4 of the peptides predicted with "SYFPEITHI".
Useful
links
For further questions, problems, etc., please, do not hesitate to contact:
Prof. Dr. Stefan Stevanovic
Technical realization by
return to top
