ElliPro: Tutorial
I. Overview
ElliPro predicts linear and discontinuous antibody epitopes based on a protein antigen's 3D structure. ElliPro accepts either a protein structure (preferred) or a protein sequence as an input. If a protein sequence is used, ElliPro will predict its 3D structure by homology modeling.
ElliPro associates each predicted epitope with a score, defined as a PI (Protrusion Index) value averaged over epitope residues. For each residue, a PI value is defined as percentage of the protein atoms enclosed in the ellipsoid, which approximates the protein surface, at which the residue first becomes lying outside the ellipsoid; for example, all residues that are outside the 90% ellipsoid will have PI = 9 (or 0.9 in ElliPro).
II. How to use the tool
- Specifying a protein sequence input
- Specifying a protein structure input
- Specifying parameters for epitope prediction
1. Input is a protein sequence
If the input is a protein sequence, ElliPro will search the PDB database for 3D structural templates using BLAST. A 3D structure will be predicted for the input protein sequence by homology modeling based on user-selected structural template(s). Linear and discontinuous epitopes will then be computed based on the predicted protein structure.
Input Page:
Step 1. Choose an input type: Select "Protein sequence", then go to Step 2a
Step 2a.
- Specify a protein sequence by entering a swiss-prot protein ID or a protein sequence (in plain or FASTA format) in the text box
- Specify "Blast expectation value" to search for sequences that have 3D structures for homology modeling. Smaller expectation value means higher sequence similarity.
- Specify maximum number of 3D structural template(s) to be returned in the search results (Note: bigger the number will make the BLAST search longer)
Step 3. Specify epitope prediction parameters (see Epitope prediction parameters)
Finally Press "Submit" (Figure 1)
Figure 1. Example sequence input using swiss-prot ID
After you press "Submit" on the input page, ElliPro performs a blast search for sequences that are similar to your input protein sequence and that have 3D structures available. Select one or more structural template(s) for homology modeling, then click "Submit" (Figure 2). (Note: selecting more structural templates will make homology modeling longer).
For good homology modeling, select template(s) that has/have:
- Long "Alignment Length"
- High number of "Identities"
- Low number of "Gaps"
Details of the sequence alignment can viewed by clicking on the "View" button in the "Alignment" column (Figure 3).
Figure 2. Example structural templates
Figure 3. Example alignment detail
2. Input is a protein structure
Input Page:
Step 1. Choose an input type: Select "Protein structure ", then go to Step 2b
Step 2b.
- Specify a protein structure by entering a protein PDB ID or uploading a protein pdb structure from file
Step 3. Specify epitope prediction parameters (see Epitope prediction parameters)
Finally Press "Submit" (Figure 4 )
Figure 4 . Example structure input using PDB ID
If your protein structure contains more than one chain, ElliPro will ask you to select which chain(s) to be used for epitope prediction. Select one or more chain(s), then click "Submit" (Figure 5 ).
Figure 5 . Example structure input using PDB ID
Important Note on Epitope Prediction:
The tool treats each protein chain independently; therefore, it might predict epitopes which are buried in the interfaces with other chains. If you wish to predict epitopes in a multi-chain protein, you need to generate a PDB file in which all chains of interest must have the same chain ID which is specified in the column 22 of the ATOM record in the PDB format file.
3. Epitope prediction parameters
- Specify "Minimum score" for epitope prediction
- Higher score will predict fewer epitopes and vice versa
- Specify "Maximum distance" for predicting (grouping) discontinous epitopes
- Longer distance will predict discontinuous epitopes that span larger regions and vice vers
III. How prediction results are presented
An example layout of prediction results is shown in Figure 6 .
The results show:
- Protein Sequence(s)
- Predicted Linear Epitope(s)
- Predicted Discontinuous Epitope(s)
Visualization
- Click on "View" in the "3D Structure" column following each predicted eptiope to view its 3D structure mapping (Figure 7 )
- Click on "Click here to view residue scores" button at the bottom of the prediction result page (Figure 6 ) to view 2D score chart(s) of the protein sequence(s) (Figure 8 )
Figure 6
