Web server usage

The TPPred3 web server is available at http://tppred3.biocomp.unibo.it. The predictor home page is shown in Figure 1

Figure 1: TPpred3 predictor home page

From the home page it is possible to perform two different actions:

  1. Start a new targeting peptide prediction job
  2. Retrieve results from a previously executed job

Start a new targeting peptide prediction

To start a new prediction job proceed as follows:

  1. Paste a single protein sequence in FASTA format into the dedicated text area
  2. Select the protein kingdom. If the protein comes from an organisms endowed with both mitochodnria and chloroplasts, please sleect "Plant". Choose "Non-plant" for any other organism.
  3. Press the “Start prediction” button to start the job

After this you will be automatically redirected to the Job Summary page shown in Figure 2.

Figure 2: Job summary page

In this page you find the main information about the current job:

  1. The assigned job id (this id will be used to retrieve this job results in the future and should be conserved)
  2. The submission and completion time of the job
  3. The current job status (this field will be automatically updated during job execution)
  4. The link to access the results (the links will be available after job completion)

As soon as the job execution is completed, the Results field will show a link “go to result page”. By clicking it, the user will be redirected to the Result page, shown in Figure 3.

Figure 3: Job result page

Retrieving results from previously executed jobs

To retrieve results of a previously executed job:

  1. Go to the TPpred3 home page
  2. Paste the job id (obtained at submission time) in the dedicated input field
  3. Click the retrieve job button

If the job ends correctly, the user will be automatically redirected to the summary page (Figure 4S) and from there it can access the job result page.


  1. Fariselli P. et al. (2009) Grammatical-restrained hidden conditional random fields for bioinformatics applications. Algorithms Mol. Biol. 22:4-13.
  2. Indio, V. et al. (2013) The prediction oforganelle-targeting peptides in eukaryotic proteins with Grammatical-Restrained Hidden Conditional Random Fields. Bioinformatics 29:981-988.