Web Application

A web app that is launched from the command line can be used to view and analyse results from a set of predictions that you have made. This is an improved and much easier to use form of a previous web interface called epitopemap and replaces it. Note: this app is still under development, suggestions for additional functionality are welcome. In the future it will probably be possible to launch predictions inside the app. For now you run predictions on the command line and view them in the browser.

Usage and Interface

After you have made some predictions you can run the app (usually from the same folder where you ran your predictions) using:

epitopepredict -s

The default port is 8888. You can use a different port by specifying it with the -x option. This should open a new web page in your browser. To view all results from a run you then enter the path (folder) where your results were saved in the form and press submit. This should refresh your form with a drop down list of all the available sequences/proteins.

There are several ways to view a set of binding predictions, all of which allow views for whichever predictors you have used. There is currently

Summary view

Summarizes the results for multiple sequences in one page. You can choose to view the table of all predicted binders, promiscuous binders or a summary over each sequence. These tables can be downloaded to csv files.

Sequence view

For viewing the detailed results for a single sequence, often representing a protein coding sequence. Graphical views of the prediction scoring across the sequence are designed to provide a quick look at the pattern of peptide binding prediction in multiple alleles. By default a track view of each allele/predictor is shown as below:


Track plots are useful for overall results over protein-length and longer sequences. The plot can be zoomed and panned using the mouse. A hover tooltip shows the particular peptide details.

Grid plots are another way to view peptide scores across a protein sequence for each allele. A hover tooltip shows the particular peptide details.



Allows you to generate a configuration file from a form for running a set of predictions. In future this could be used to submit jobs directly.

Future features

  • Improved graphical features for genome based prediction.
  • Location of clusters of binders in sequences.
  • Export of peptide lists/n-mers for experimental use.
  • Mutation/conservation analysis.
  • Edit config and run predictions from web page.