Events News Research CBS CBS Publications Bioinformatics
Staff Contact About Internal CBS CBS Other

NetMHC 3.0 Server

NEW UPDATED VERSION: NetMHC-3.2

STAND ALONE VERSION: 3.0 Download for academics

NetMHC 3.0 server predicts binding of peptides to a number of different HLA alleles using artificial neural networks (ANNs) and weight matrices.

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

Predictions can be obtained for 12 human supertypes including more than 100 individual human alleles using ANNs and PSSMs (ungapped HMMs). Furthermore 16 animal (Monkey and Mouse) allele predictions are available. ANNs have been trained for 43 different Human MHC (HLA) alleles representing all 12 HLA A and B Supertypes as defined by Lund et al. (2004). Weight matrices are generated using an ungapped HMM approach as described in Nielsen et al. (2004) with data from the SYFPEITHI database.

For ANN prediction values are given in nM IC50 values. For weight matrices prediction values are given as a fitness score, so that a high fitness score correlates to strong binding.

Predictions of lengths 8-11:       Predictions can be made for lengths between 8 and 11 for all alleles using an novel approximation algorithm using ANNs trained on 9mer peptides. Probably because of the limited amount of available 10mer data this method has a better predictive value than ANNs trained on 10mer data. However, caution should be taken for 8mer predictions as some alleles might not bind 8mers to any significant extend.

For both ANN and weight matrix predictions strong and weak binding peptides are indicated in the output. In the selection window for HLA alleles, the recommended allele for each HLA supertype is indicated.

The project is a collaboration between CBS and IMMI.

 

Instructions Output format Article abstract Evaluation Set

SUBMISSION

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

Optionally paste a number of peptides AND select the peptide input checkbox:

Submit a file in FASTA format OR raw peptide format (check "Peptide input") directly from your local disk:

 

Peptide input 

MHC Alleles                  Peptide length   

Sort by affinity 

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

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


CITATIONS

For publication of results, please cite:

  • Sensitive quantitative predictions of peptide-MHC binding by a 'Query by Committee' artificial neural network approach.
    Buus S, Lauemoller SL, Worning P, Kesmir C, Frimurer T, Corbet S, Fomsgaard A, Hilden J,Holm A, Brunak S.
    Tissue Antigens., 62:378-84, 2003.

    View the abstract or the full text version at PMID: 14617044

  • Reliable prediction of T-cell epitopes using neural networks with novel sequence representations.
    Nielsen M, Lundegaard C, Worning P, Lauemoller SL, Lamberth K, Buus S, Brunak S, Lund O.
    Protein Sci., 12:1007-17, 2003.

    View the abstract or the full text version at PMID: 12717023

  • If peptide lengths other than 9mers were predicted and the approximation method were used (see output) please cite:

    Accurate approximation method for prediction of class I MHC affinities for peptides of length 8, 10 and 11 using prediction tools trained on 9mers.
    Lundegaard C, Lund O, Nielsen M.
    Bioinformatics, 24(11):1397-98, 2008.

    Get abstract and link to full text at NCBI PubMed with PMID 18413329.

  • If you specifically use the weight matrix derived predictions, please also cite:

    Improved prediction of MHC class I and class II epitopes using a novel Gibbs sampling approach.
    Nielsen M, Lundegaard C, Worning P, Hvid CS, Lamberth K, Buus S, Brunak S, Lund O.
    Bioinformatics, 20(9):1388-97, 2004.

    View the abstract or the full text version at PMID: 14962912


PORTABLE VERSION

Would you prefer to run NetMHC at your own site? NetMHC v. 3.0 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 software@cbs.dtu.dk.


GETTING HELP

Scientific problems:        Technical problems: