Services
NetMHCcons - 1.1
Binding of peptides to any known MHC class I molecule
The NetMHCcons 1.1 server predicts binding of peptides to any known MHC class I molecule. This is a consensus method for MHC class I predictions integrating three state-of-the-art methods NetMHC, NetMHCpan and PickPocket to give the most accurate predictions. The server also allows to use each of these methods separately:
NetMHC is an artificial neural network-based (ANN) allele-specific method which has been trained using 94 MHC class I alleles. Version 3.4 is used as part of NetMHCcons-1.1.
NetMHCpan is a pan-specific ANN method trained on more than 115,000 quantitative binding data covering more than 120 different MHC molecules. Version 2.8 is used as part of NetMHCcons-1.1.
PickPocket method is matrix-based and relies on receptor-pocket similarities between MHC molecules. It has been trained on 94 different MHC alleles. In the PickPocket version 1.1, the matrices of pocket-library are generated using the SMMPMBEC method.
NetMHCcons 1.1 server can produce predictions for peptides of 8-15 amino acids in length. It also gives a possibility to choose several lengths.
Two submission types are handled - the list of peptides or a protein sequence in FASTA format. The server provides a possibility for the user to choose MHC molecule in question from a long list of alleles or alternatively upload the MHC protein sequence of interest.
The prediction values are given in nM IC50 values and as % Rank to a set of 200.000 random natural peptides. The user has a choice of setting the threshold for defining strong and weak binders based on predicted affinity (IC50) or % Rank. Strong and weak binding peptides will be indicated in the output. The output can also be sorted based on predicted binding affinity as well as filtered on the user-specified thresholds.
The project is a collaboration between CBS, IMMI at Copenhagen university and LIAI.
Submission
CITATIONS
For publication of results, please cite:
-
NetMHCcons: a consensus method for the major histocompatibility complex class I predictions
Edita Karosiene, Claus Lundegaard, Ole Lund and Morten Nielsen
PMID: 22009319
Full text
Other relevant publications:
-
NetMHC-3.0: accurate web accessible predictions of human, mouse and monkey MH
C class I affinities for peptides of length 8-11
Lundegaard C, Lamberth K, Harndahl M, Buus S, Lund O, Nielsen M.
Nucleic Acids Res. 1;36(Web Serve r issue):W509-12. 2008View the abstract.
-
NetMHCpan - MHC class I binding prediction beyond humans
Ilka Hoof, Bjoern Peters, John Sidney, Lasse Eggers Pedersen, Ole Lund, Soren Bu us, and Morten Nielsen
PMID: 19002680 Full text -
The PickPocket method for predicting binding specificities for receptors based on receptor pocket similarities: application to MHC-peptide binding
Zhang H, Lund O, Nielsen M. Bioinformatics. 2009 May 15;25(10):1293-9
DATA RESOURCES
Data resources used to develop this server was obtained from
- IEDB database.
- Quantitative peptide binding data were obtained from the IEDB database.
Instructions
1. Specify the input sequences
All the input sequences must be in one-letter amino acid code. The allowed alphabet (not case sensitive) is as follows:
All the other symbols will be converted to X before processing.
The server allows for input in either FASTA or PEPTIDE format.
Note that for Peptide input, all peptides MUST be of equal length.
The sequences can be input in the following two ways:
-
Paste a single sequence (just the amino acids) or a number of sequences in
FASTA
format or a list of peptides into the upper window of the main server page.
- Select a FASTA or PEPTIDE file on your local disk, either by typing the file name into the lower window or by browsing the disk.
Both ways can be employed at the same time: all the specified sequences will
be processed. However, there may be not more than 500 sequences
in total in one submission. The sequences shorter than 15
or longer than 10000 amino acids will be ignored.
2. Customize your run
1. Select peptide lenght from the list. Multiple lengths can be chosen by using Ctrl key. If no lenght will be selected, the method will give predictions for 9mer peptides.
2. Select a method which you want to use to get predictions. NetMHCcons is a default method, which integrates all 3 methods and gives the best predictions for the query allele. But if needed NetMHC, NetMHCpan or PickPocket methods can be used separately.
3. Select the allele(s) you want to make predictions for from the scroll-down menu or type in the allele names separated by commas (with out blank spaces). Note that for NetMHC method not all the alleles are available. When you choose NetMHC method, the lists of possible alleles to choose will be updated. If you want to type in the alleles, check the list of alleles before choosing the query allele if you want to use NetMHC method.
4. You can also paste a single full length MHC protein sequence in FASTA format or submit a file containing a full length MHC protein sequence in FASTA format directly from your local disk.
5. Specify thresholds for strong and weak binders. Two different types of thresholds can be set: based on the binding affinity giving in nM IC50 values or based on % Rank obtained using the method on 200.000 random natural peptides. The peptide will be identified as a strong binder if the % Rank OR binding affinity (IC50) is below the specified threshold for the strong binders. The peptide will be identified as a weak binder if the % Rank OR binding affinity (IC50) is above the threshold of the strong binders but below the specified threshold for the weak binders.
6. Choose if you want to filter the output so that not all the predictions will be shown. If you choose Yes, you will be able to choose filtering thresholds for % Rank and IC50 values. Only predictions for peptides that will have the % Rank OR binding affinity (IC50) below the specified thresholds will be shown in the output.
7. Click the box Sort by affinity to have the output sorted by descending predicted binding affinity.
8. Click the box Save prediction to xls file if you want the output to be saved in xls file.
3. Submit the job
Click on the "Submit" button. The status of your job (either 'queued' or 'running') will be displayed and constantly updated until it terminates and the server output appears in the browser window.At any time during the wait you may enter your e-mail address and simply leave the window. Your job will continue; you will be notified by e-mail when it has terminated. The e-mail message will contain the URL under which the results are stored; they will remain on the server for 24 hours for you to collect them.
Format of NetMHCcons-1.1 output
DESCRIPTION
The output provides common information for all the alleles:
For each allele the allele name is given and if NetMHCcons method was chosen, it is specified which method or combination of methods was used to get predictions for that allele.
The prediction output for each allele consists of 8 columns:
Fianlly, at the end of predictions for each allele, the numbers of total strong and weak binders are given.
EXAMPLE OUTPUT
# Method: NetMHCcons
# Input is in FASTA format
# Peptide length 9
# Threshold for Strong binding peptides (IC50) 50.000 nM
# Threshold for Weak binding peptides (IC50) 500.000 nM
# Threshold for Strong binding peptides (%Rank) 0.5%
# Threshold for Weak binding peptides (%Rank) 2%
# Allele: HLA-A01:01
# NetMHCcons = NetMHC+NetMHCpan
---------------------------------------------------------------------------------------
pos Allele peptide Identity 1-log50k(aff) Affinity(nM) %Rank BindingLevel
---------------------------------------------------------------------------------------
0 HLA-A*01:01 ASTPGHTII seq1 0.092 18478.42 15.00
1 HLA-A*01:01 STPGHTIIY seq1 0.358 1044.93 0.80 <=WB
2 HLA-A*01:01 TPGHTIIYE seq1 0.038 32965.50 50.00
3 HLA-A*01:01 PGHTIIYEA seq1 0.044 30893.41 50.00
4 HLA-A*01:01 GHTIIYEAV seq1 0.057 27131.77 50.00
5 HLA-A*01:01 HTIIYEAVC seq1 0.062 25564.30 32.00
6 HLA-A*01:01 TIIYEAVCL seq1 0.057 27131.77 50.00
7 HLA-A*01:01 IIYEAVCLH seq1 0.069 23571.74 32.00
8 HLA-A*01:01 IYEAVCLHN seq1 0.046 30396.07 50.00
9 HLA-A*01:01 YEAVCLHND seq1 0.037 33686.63 50.00
10 HLA-A*01:01 EAVCLHNDR seq1 0.046 30232.07 50.00
11 HLA-A*01:01 AVCLHNDRT seq1 0.050 29266.51 50.00
12 HLA-A*01:01 VCLHNDRTT seq1 0.036 34053.09 50.00
13 HLA-A*01:01 CLHNDRTTI seq1 0.061 25842.40 32.00
14 HLA-A*01:01 LHNDRTTIP seq1 0.039 32787.64 50.00
----------------------------------------------------------------------------------------
Number of strong binders: 0 Number of weak binders: 1
----------------------------------------------------------------------------------------
References
Main reference:
NetMHCcons: a consensus method for the major histocompatibility complex class I predictions
Edita Karosiene, Claus Lundegaard, Ole Lund, and Morten Nielsen.
Immunogenetics, 2011
Center for Biological Sequence Analysis,
Technical University of Denmark,
DK-2800 Lyngby, Denmark
A key role in cell-mediated immunity is dedicated to the major histocompatibility complex (MHC) molecules that bind peptides for presentation on the cell surface. Several in silico methods capable of predicting peptide binding to MHC class I have been developed. The accuracy of these methods depends on the data available characterizing the binding specificity of MHC molecules. It has moreover been demonstrated that consensus methods defined as combinations of two or more different methods led to improved prediction accuracy. This plethora of methods makes it very difficult for the non-expert user to choose the most suitable method for predicting binding to a given MHC molecule. In this study, we have therefore made an in-depth analysis of combinations of three state-of-the-art MHC-peptide binding prediction methods (NetMHC, NetMHCpan and PickPocket). We demonstrate that a simple combination of NetMHC and NetMHCpan gives the highest performance when the allele in question is included in the training and is characterized by at least 50 data points with at least 10 binders. Otherwise, NetMHCpan is the best predictor. When an allele has not been characterized, the performance depends on the distance to the training data. NetMHCpan has the highest performance when close neighbours are present in the training set, while the combination of NetMHCpan and PickPocket outperforms either of the two methods for alleles with more remote neighbours. The final method, NetMHCcons, is publicly available at www.cbs.dtu.dk/services/NetMHCcons, and allows the user in an automatic manner to obtain the most accurate predictions for any given MHC molecule.
PMID: 22009319
Full text