dataset.zip

Our final dataset comprises of 491,018 class I MHC allele data points covering 161 different HLAs and 64,954 class II MHC allele data points covering 49 different HLAs. The peptide lengths range from 3 to 43 for both classes, while the lengths of amino acid sequences of HLAs range from 180 to 347 for class I and from 85 to 232 for class II.

We get 451,484 HLA-A*, 39,424 HLA-B* and 110 HLA-C* class I alleles, out of which 379,783 are binding and 111,235 are non-binding. Similarly, we get 64,926 HLA-DRB1*, 22 HLA-DRB3*, 4 HLA-DRB4* and 2 HLA-DRB5* class II alleles out of which 36,035 are binding and 28,919 are non-binding.