Description
The subtracks of this track shows peptides that are predicted to be displayed by
human and macaque class I MHCs and human class II MHCs. Class I predictions
are split by HLA allele, class II predictions are summarized into a single track.
Display Conventions and Configuration
Class I peptides are shaded by IC50 values and grouped by HLA-alleles. Class II
peptides are shaded by percentile.
Methods
Class I T cell epitope prediction:
For human epitope prediction, binding affinity for all possible 9-mer peptides
from each protein sequence to a set of 27 HLA alleles that are most frequent in
the global population, using the SMM method available in the IEDB MHC class
I binding prediction tool. All peptides with predicted binding affinity (expressed
as IC50 nM) below previously optimized allele-specific thresholds were
selected as predicted binders. In the case of rhesus macaque class I T cell
epitope prediction, the binding affinity of all 9-mer peptides was predicted
for a total of 19 alleles (7 Indian and 12 Chinese rhesus macaques) that are
most frequent in these species. The binding affinity predictions utilized the
"IEDB recommended" method from the IEDB MHC class I binding prediction tool.
Peptides with the IEDB consensus percentile rank <= 1.0 were selected. Lower
values for either IC50 or percentile rank indicate stronger binding.
Class II T cell epitope prediction: For human epitope predictions, 15-mer
peptides overlapping by 10 aa residues were generated from aligned sequences,
to avoid redundant peptides that share the same 9-mer binding core.
For each 15-mer peptide, the binding affinity was predicted (expressed
as the IEDB consensus percentile rank) for seven class II human HLA alleles
(HLA-DRB1*03:01, HLA-DRB1*07:01, HLA-DRB1*15:01, HLA-DRB3*01:01,
HLA-DRB3*02:02, HLADRB4*01:01 and HLA-DRB5*01:01), using the IEDB MHC class
I binding prediction tool ("IEDB recommended" method). Predicted binders were
selected based on the median consensus percentile rank estimated from the
consensus percentile ranks for the seven alleles. All peptides with median
consensus percentile rank <= 20.0 were selected as predicted binders, the
threshold being previously optimized for capturing 50% of class II human T cell
responses.
|
|