Services are gradually being migrated to
Please try out the new site.

NetMHCII 2.2 Server

NetMHCII 2.2 server predicts binding of peptides to HLA-DR, HLA-DQ, HLA-DP and mouse MHC class II alleles using articial neuron networks.

Predictions can be obtained for 14 HLA-DR alleles covering the 9 HLA-DR supertypes, six HLA-DQ, six HLA-DP, and two mouse H2 class II alleles.

The prediction values are given in nM IC50 values, and as a %-Rank to a set of 1,000,000 random natural peptides. Strong and weak binding peptides are indicated in the output.

Update as of 29/9/2011. The accuracy of the peptide binding core identification has been improved using a neural network alignment procedure. This has made obsolete the need for P1 amino acids encoding. Recomended use of the server is hence without P1 encode. This is now the default setting of the method and gives optimal performance both in term of binding core identification and binding affinity prediction.

Update as of 29/11/2011. The allele naming for HLA-DPA10301-DPB10401 was wrong. The typing of this molecule is incomplete. The beta chain is either DPB10301 or DPB10401. The alpha chain is DPA10103. This is now reflected in the name used for this allele (HLA-DPA10103-HLA-DPB10301_DPB10401).

Update as of 03/02/2012. The accuracy of the peptide binding core identification has been improved using a neural network alignment procedure for all alleles. The network alignment procedure is described in PMID. This procedure does NOT affect the predicted binding values.

View the version history of this server. All the previous versions are available on line, for comparison and reference.

Instructions Output format Article abstract Training and Evaluation Data


Type of input

Paste a single sequence or several sequences in FASTA format into the field below:

or submit a file in FASTA format directly from your local disk:

Peptide length  

Select Loci

Select Allele (max 15 per submission) or type allele names (ie DRBX_XXXX) seperated by commas (max 15 per submission)  

Threshold         Sort by affinity 

Turn on P1 amino acid preference  

At most 5000 sequences per submission; each sequence not more than 20,000 amino acids and not less than 9 amino acids.

The sequences are kept confidential and will be deleted after processing.


For publication of results, please cite:

  • NN-align. A neural network-based alignment algorithm for MHC class II peptide binding prediction.
    Nielsen M and Lund O
    PMID: 19765293
    Full text
  • Prediction of MHC class II binding affinity using SMM-align, a novel stabilization matrix alignment method.
    Nielsen M, Lundegaard C, Lund O.
    PMID: 17608956
    Full text


Would you prefer to run NetMHCII at your own site? NetMHCII v. 2.2 is available as a stand-alone software package, with the same functionality as the service above. Ready-to-ship packages exist for the most common UNIX platforms. There is a download page for academic users; other users are requested to contact CBS Software Package Manager at


Scientific problems:        Technical problems: