Skip to main content

Table 2 Details of the benchmark calculation covering the 14 HLA-DR alleles.

From: Prediction of MHC class II binding affinity using SMM-align, a novel stabilization matrix alignment method

AUC
Allele SMM Gibbs TEPITOPE SVRMHC MHCpred ARB SMM -PRF NetMHCII N
1*0101 0.702 0.676 0.647 0.623 0.565 0.666 0.716 0.716 1203
1*0301 0.779 0.722 0.734    0.799 0.770 0.765 474
1*0401 0.741 0.759 0.754 0.739 0.606 0.737 0.756 0.758 457
1*0404 0.798 0.743 0.829    0.788 0.808 0.785 168
1*0405 0.727 0.724 0.790 0.701   0.724 0.733 0.735 171
1*0701 0.768 0.695 0.768   0.647 0.749 0.774 0.787 310
1*0802 0.724 0.721 0.769    0.803 0.740 0.756 174
1*0901 0.726 0.734     0.711 0.759 0.775 117
1*1101 0.715 0.715 0.710    0.727 0.720 0.734 359
1*1302 0.810 0.716 0.720    0.917 0.819 0.818 179
1*1501 0.715 0.672 0.726 0.730   0.792 0.733 0.736 365
3*0101 0.620 0.512     0.717 0.771 0.815 102
4*0101 0.730 0.742     0.800 0.729 0.736 181
5*0101 0.664 0.618 0.653 0.649   0.677 0.655 0.664 343
  1. The predictive performance is shown in terms of the area under the ROC curve (AUC) for the SMM-align, Gibbs sampler [4], TEPITOPE [3], SVRMHC [7], MHCpred [15], and ARB methods, respectively. The SMM-PRF method refers to the extended SMM align method including penalties for long peptides and short amino terminal peptide flanking residues, and the NetMHCII method refers to the final extended SMM align method including direct encoding of peptide flanking residues and penalties for longer peptides and short amino terminal peptide flanking residues. The first column gives the allele names as 1*0101 for DRB1*0101 etc The last column gives the number of peptide data included for each allele. For each allele, the performance of the SMM-align, Gibbs sampler, and NetMHCII methods was estimated using five-fold cross-validation as described in the text. The details of the benchmark calculation as measured in terms of the Pearson's and Spearman's rank correlation are shown in Supplementary materials table 1 [see Additional file 1].