Prediction of T cell epitopes


During this exercise you will use bioinformatics tools to predict peptide-MHC binding. The exercise has three parts:

  1. Identification of MHC binding motif
  2. Visualize the motif using sequence logos
  3. Training of MHC class I and class II binding prediction methods

Background: Peptide MHC binding

The most selective step in identifying potential peptide immunogens is the binding of the peptide to the MHC complex. Only one in about 200 peptides will bind to a given MHC complex. A very large number of different MHC alleles exist each with a highly selective peptide binding specificity.

The binding motif for a given MHC class I complex is in most cases 9 amino acids long. The motif is characterized by a strong amino acid preference at specific positions in the motif. These position are called anchor positions. For many MHC complexes the anchor position are placed at P2 and P9 in the motif. However this is not always the case.

Large number of peptide data exist describing this MHC specificity variation. One important source of data is the SYFPEITHI MHC database ( This database contains information on MHC ligands and binding motifs.

Purpose of exercise, description of data

In this exercise you are going to

  1. Search the SYFPEITHI database to characterize the binding motif for different MHC alleles.
  2. Visualize the binding motif using sequence logos.
  3. Use the Easypred and EasyGibbs web-interfaces to train bioinformatics predictor for MHC-peptide binding.

The exercise

Identification of MHC binding motifs

Go to the SYFPEITHI MHC database (click on the right mouse button, and select open in a new window). Use the Find motif, Ligand or epitope option. Have a look at the peptide characteristics for different MHC alleles like HLA-A*0201, HLA-A*01, HLA-A*1101, and HLA-B27. This you do by selecting for instance HLA-A*0201 and press do Query.

  • Q1: What are the characteristics of the peptides that bind HLA-A*0201, that is which positions are anchor position and what amino acids are found at the anchor positions?
Repeat the analysis for a few (at least two) other MHC alleles.
  • Q2: Does the location of anchors differ, and does the amino acid preferences differ between different MHC alleles?

Sequence logos

A powerful way to visualize the peptide characteristics of the binding motif of an MHC complex, is to plot a sequence logo. The files HLA-A01, HLA-A0201, HLA-B27, in the exercise directory contain peptides known to bind to a particular MHC complex (HLA-A*0201 for example). The files are in the format used by a program that generates sequence logos. Go to the web-site EasyPred.

This webserver can do a lot of fun stuff. For now you shall just use it to visualize sequence logos. This you by pasting in the three files one at the time into the training examples window and type Submit query. Do this for each of the three peptide files and examine the sequence logos. Do not worry abot all he other output. We will explain this later.

  • Q3: Does the anchor positions and amino acid preferences correspond to what you found using the SYFPEITHI database?

Prediction of MHC-peptide binding

In this part of the exercise you shall use the EasyPred web-interface to train and evaluate a series of different MHC-peptide binding predictors. You shall use two data sets (eval.set, train.set) that contain peptides and binding affinity to the MHC alleles HLA-A*0201. The binding affinity is a number between 0 and 1, where a high value indicates strong binding (a value of 0.5 corresponds to a binding affinity of approximately 200 nM). The eval.set contains 66, and the train.set 1200 such peptides. Click on the filenames to view the content of the files.

Before you start using the EasyPred you must save the train.set and eval.set files locally on the Desktop on your lab-top. You do that by clicking on the files names (eval.set, train.set) and saving the files as text files on the Desktop.

You shall now use EasyPred web-server to train a series of methods to predict peptide-MHC binding. Go to the EasyPred web-server.

Weight Matrix construction

First you shall train a matrix predictor. On the EasyPred web-server press Clear fields. In the upload training examples window browse and select the train.set file from the Desktop, in the upload evaluation window browse and select the eval.set file from the Desktop. In the Matrix method parameters select Clustering at 62% identity, and set weight on prion (weight on pseudo counts) to 200. Press Submit query. This will calculate a weight-matrix using sequence weighting by clustering, and a weight on prior (pseudo counts) of 200.

  • Q4: What is the predictive performance of the matrix method (Pearson coefficient and Aroc value)?
  • Q5: How many of the 1200 peptides in the train set are included in the matrix construction?

Go back to the EasyPred server window (use the Back bottom). Set clustering method to No clustering and the weight on prior to zero and redo calculation.

  • Q6: What is the predictive performance of the matrix method now?
  • Q7: Does clustering and pseudo count (weight on prior) improve the prediction accuracy?

Neural networks

Training set partition

Now the fun starts. You shall now train some neural networks to predict MHC-peptide binding. In the Type of prediction method window select neural networks. Leave all other parameters as they are. Press Submit query.

This will train a neural network with 2 hidden neurons running up-to 300 training epochs. The top 80% (960 peptides) of the train.set is used to train the neural network and the bottom 20% (240 peptides) are used to stop the training to avoid over-fitting.

  • Q8: What is the maximal test performance (maximal test set Pearson correlation), and in what epoch (training cycle) does it occur?
  • Q9: What is evaluation performance (Pearson correlation and Aroc values)?

Go back to the EasyPred interface and change the parameters so that you use the bottom 80% of the train.set to train the neural network and the top 20% to stop the training. Redo the network training with the new parameters.

  • Q10: What is the maximal test performance, and in what epoch does it occur?
  • Q11: What is evaluation performance?
  • Q12: How does the performance differ from what you found in the previous training?
  • Q13: Why do you think the performance differ so much?

Hidden Neurons

Go back to the EasyPred interface and change the parameters back so that you use the top 80% of the train.set of training. Next do neural network training with a different set of hidden neurons (1 and 5 for instance).

  • Q14: How does the test performance differ when you vary the number of hidden neurons?
  • Q15: How does the evaluation performance differ?
  • Q16: Can you decide on an optimal number of hidden neuron?
  • Q17: Why do you think the number of hidden neurons has so little importance?

Cross-validated training

As you found in the first part of neural network training, the network performance can depend strongly on the partition of the training data into the training and stop set. One way of improving the network performance is to make use of this network variation in a cross-validated training. The general idea behind the cross-validated training is that since you cannot in advance tell which training set partition that will be optimal you make a series of N network trainings each with a different partition. The final network prediction is then taken as the simple average over the N predictions. In a 5-fold cross-validated training, the training set is split up into 5 sets. In one training the sets 1,2,3 and 4 are used to train the network and the 5th set to stop the training, in the another training the sets 1,3,4,5 are used for training and the 2nd set to stop the training, and so forth.

Go back to the EasyPred interface and set the hidden neuron parameter back to 2. Next set the number of partitions for cross-validated training to 5 and redo the neural network training (this might take some minutes).

Write down the test performance for each of the five networks

  • Q18: How does the train/test performance differ between the different partitions?
  • Q19: What is the evaluation performance and how does it compare to the performance you found previously?

Now you must save the parameters for the cross-validated network you have just trained. Use the right mouse-bottom on the Parameters for prediction method to save the neural network parameters to a file (say para.dat). You can now run predictions using this neural network without redoing network training by uploading the parameter file in the Load saved prediction method window.

Finding epitopes in real proteins

You shall use the neural network to find potential epitopes in the Sars virus. In the EasyPred web-interface clear field to reset all parameter fields. Go to the Swiss-prot homepage Swiss-prot. Search for a Sars entry by typing Sars in the search window. Click you way to the FASTA format for one of the proteins. Here is a link if you are lazy. Paste in FASTA file into the Paste in evaluation examples. Upload the network parameter file (para.dat) from before into the Load saved prediction method window. Leave the window Networks to chose in ensemble blank, make sure that the option for sorting the output is set to Sort output on predicted values, and press Submit query.

  • Q20: How many high binding epitopes do you find? Is this number reasonable (how large a fraction of random 9meric peptides are expected to bind to a given HLA complex?)

Training a prediction method for HLA class II binding.

If you have more time, you shall now make a prediction method for MHC class II binding.

HLA class II binding peptides have a broad length distribution complicating the development of prediction methods. Identifying the correct alignment of a set of peptides known to bind the MHC class II complex is a crucial part of identifying the core of an MHC class II binding motif.

Here you shall use the EasyGibbs web-server to develop and evaluate a prediction method for the HLA class II allele DR*0401. The data used to train the method are downloaded from the SYFPEITHI MHC database, and the data used to evaluate the predictive performance are taken from a paper by Southwood et al. 1998. The ClassII_train.set contains 277 peptides known to bind to the allele DR*0401, and the ClassII_eval.set contains 22 peptides with associated binding affinity.

Go to the EasyGibbs webpage. In the upload evaluation window browse and select the ClassII_eval.set file from the Desktop. Do not upload any training file!!. In the window Load saved prediction method paste in the content of the file dr1_0401.mat. This will upload the very popular prediction matrix of the TEPITOPE method. In the window Cutoff for counting an example as a positive example. set the value to 0.362 (this corresponds to close to 1000 nM). Next press Submit query.

  • Q21: What is the predictive performance of the method?
Next close the EasyGibbs window and open it again (some how the Clear fields does not always work) EasyGibbs. In the upload training examples window browse and select the ClassII_train.set file from the Desktop, in the upload evaluation window browse and select the ClassII_eval.set file from the Desktop. In the window Cutoff for counting an example as a positive example. set the value to 0.362 (this corresponds to close to 1000 nM). Next press Submit query. This will start the Gibbs sampler to search for a 9mer binding motif in the training data, and next apply the identified motif to predict binding for the peptides in the evaluation set.

  • Q22: What is the predictive performance of the method?

Now go back to the EasyGibbs interface and change the clustering method to Cluster at 62% identity, and set Weight on prior to 200. Next press Submit query.

From the information on the SYFPEITHI webpage one can see that the most conserved position in the binding motif is P1 (only 7 allowed amino acids). We can use this information to improve the prediction accuracy.

Go back to the EasyGibbs interface and in the window Position specific weighting set the weigh on position P1 to 2. This is done by typing 2,1,1,1,1,1,1,1,1 in the window. Leave all other setting unchanged. Press Submit query.

  • Q26: What is the predictive performance of the method? and how does the performance compare to that of the TEPITOPE method?

Now you have within less than 3 hours developed advanced and competitive methods for predicting binding of peptides to HLA class I and II. Also you have identified potential CTL epitope vaccine candidates for the SARS virus. All you need now is to find some venture capital and make your own Biotec startup company. That is not so bad an outcome of 3 hours work!

Now you are done!!