Computational protocol: Helper T Cell Epitope-Mapping Reveals MHC-Peptide Binding Affinities That Correlate with T Helper Cell Responses to Pneumococcal Surface Protein A

Similar protocols

Protocol publication

[…] IEDB (http://www.immuneepitope.org/), SYFPEITHI (http://www.syfpeithi.de/), SVMHC (http://www.bs.informatik.unituebingen.de/SVMHC/), RANKPEP (http://bio.dfci.harvard.edu/RANKPEP/), and MHCPred (http://www.jenner.ac.uk/MHCPred) external software(s) were used to predict peptide binding affinities to mouse I-A and I-E as well as HLA-DR, -DP and -DQ. In brief, for average relative binding (ARB) evaluation, 10-fold cross validation results stored at IEDB were used to estimate performance. Because the binding of peptides to MHC class II molecules is not dependent on exact size, derivation of MHC class II ARB matrices followed an iterative procedure. For the first iterative step, a matrix was generated from a set of nine-residue core sequences randomly obtained from each peptide sequence in the training set. For subsequent cycles, nine-residue core sequences were used to generate a matrix. The overall binding affinity of a peptide was predicted using the highest scoring nine-residue core sequence. For the SYFPEITHI prediction, we patched each testing peptide with three glycine residues at both ends before evaluation for prediction. This was recommended by the creators of SYFPEITHI method to ensure that all potential binders were correctly presented to the prediction algorithm. For all other methods, the original tested peptides were submitted directly for prediction. Peptide sequences were sent to web servers one at a time and predictions were extracted from the server's response. To assign a single prediction for peptides longer than nine amino acids in the context of tools predicting the affinity of 9 core-binding regions, we took the highest affinity prediction of all possible 9-mers within the longer peptide as the prediction result. For each MHC class II molecule whose binding can be predicted by three or more algorithms, the top three methods were selected that gave the best performance. For each method, peptides were tested and ranked by their scores with higher ranks for better binders. For each tested peptide, three ranks from different methods were taken and the median rank was taken as the consensus score. Peptides were classified into binders (IC50<500 nM) and nonbinders (IC50≥500 nM), as practical cutoffs. […]

Pipeline specifications

Software tools SVMHC, RANKPEP, MHCPred, ARB
Databases SYFPEITHI IEDB
Application Immune system analysis
Organisms Mus musculus
Diseases Pneumococcal Infections, Pneumonia, Leukemia, T-Cell