Welcome to SHARP2. This is a new web server for the prediction of protein interaction sites on the surface of 3D protein structures. A series of residues patches are defined and analysed for six parameters; S:Solvation Potential,H:Hydrophobicity, A:Accessible Surface Area, R:Residue Interface Propensity, P:Protrusion, P:Planarity. A combined parameter score is calculated for each patch and the patches with the highest scores are predicted to be protein interaction sites.

  ¤ SHARP2 Server
  ¤ Introduction
  ¤ Help Page
  ¤ References

Help Page

 How to Use
Step 1 - Set up basic information for a prediction
Step 2 - Select the type of your protein complex
Step 3 - Start SHARP2
Option A - Set up each parameter manually
Option B - Set up multiple predictions
Option C - Use your own coordinates file
Option D - Receive the result of your prediction by Email
Appendix - Example of result pages
  • Step1: Set up basic information for a prediction



    1. Enter the PDB entry ID
      The PDB entry ID for atomic coordinates of the crystal structures of a protein complex in which you are interested is entered in the first entry box. The PDB ID is represented by a four alphanumeric identifier, e.g 1cdt. If you do not know a PDB ID, search the Protein Data Bank (PDB) .

    2. Enter the chain ID
      The chain identifier for the protomer of choice in the PDB file selected. The chain ID is represented by a single alphanumeric identifier. (e.g A or B)



  • Step2: Select protein type



    If you select the type of your interesting protein complex, the chemical and biophysical parameters will be automatically set up according to the type of the protein, but it is possible to change the default settings, See Option A.

    Also, the patch size, which is the number of amino acids defined in a surface patch, will be set to default value associated with the type of protein complex, but it is possible to change the defalut value. more information, please see Introduction or Jones & Thornton (1997(b)).

    1. Interacting partner is identical protein (e.g homodimer)
      Parameter State Level
      Solvation Potential : on : high
      Hydrophobicity : on : high
      Accessible Surface Area : on : high
      Residue Interface Propensity : on : high
      Planarity : off : -
      Protrusion : off : -


    2. Interacting partner is different protein that is larger (e.g inhibitor)
      Parameter State Level
      Solvation Potential : on : high
      Hydrophobicity : on : high
      Accessible Surface Area : on : high
      Residue Interface Propensity : on : high
      Planarity : off : -
      Protrusion : off : -


    3. Interacting partner is different protein that is smaller (e.g enzyme)
      Parameter State Level
      Solvation Potential : off : -
      Hydrophobicity : off : -
      Accessible Surface Area : on : high
      Residue Interface Propensity : on : high
      Planarity : on : low
      Protrusion : on : high


    4. Interacting partner is an antibody (e.g antigen)
      Parameter State Level
      Solvation Potential : on : low
      Hydrophobicity : on : low
      Accessible Surface Area : on : high
      Residue Interface Propensity : off : -
      Planarity : on : high
      Protrusion : on : high




  • Step3: Start SHARP2


    The number of best patches, which are defined as the first N number of patches in descending order of the probability of forming interaction sites. A default value of 10 will display the 10 "best" patches (i.e. those most likely to be interaction sites) for any prediction, but it is possible to change the default setting.

    If you have entered all the essential details and parameters for your prediction click "SHARP2 ! " button to submit the prediction to the server.

  • Option A: Set up each parameter manually



    Instead of the automatic parameter setting, the SHARP2 allows users to define their own "best patches" enabling the inclusion or exclusion of any of the six parameters.

    1. STATE : Select on (= inclusion) or off (= exclusion) for each parameter.



    2. LEVEL : Select high (= positive) or low (= negative) for each parameter.
      Hence to select for the most hydrophobic patches on the surface of a protein you set the hydrophobicity parameter STATE to 'on' and the LEVEL to 'high', and set the remaining five parameter STATEs to OFF.



  • Option B: Set up multiple predictions (Batch submission)



    The SHARP2 allows more than one prediction with different parameter settings to be made with a single submission. (Currently the limit for multiple submissions is 5 predictions). If you want batch submissions of multiple predictions, select "Set up multiple parameter sets". Then, click "Load", and the details for each prediction will be entered into the text window (if Steps 1 and 2 have been completed) using the SHARP2 Syntax. Alternatively you can directly paste the information into the text field in the correct syntax.



    ♠ SHARP2 Syntax

    > PDBID ChainID PatchSize BestCF Ssp(s:l) Shy(s:l) Sasa(s:l) Srp(s:l) Spi(s:l) Spl(s:l)
    (1)(2)(3)(4)(5)(6) s=state, l=level


    (1) > the beginning of new line must be begun with " > ".
    (2) PDBID PDB entry identifier (e.g 1cdt)
    (3) ChainID chain identifier (e.g A, B)
    (4) PatchSize the size of a patch
    (5) BestCF the number of best patches
    (6) Sx(s:l) parameter setting
    x (parameter name) = sp(Solvation Potential), hy(Hydrophobicity), asa(Accessible Surface Area), rp(Residue Interface Propensity), pi(Protrusion) or pl(Planarity).
    s (state) = 1(=on) or 0(=off)
    l (level) = 1(=hight) or 0(=low)


  • Option C: Use your own coordinates file


    The server allows users to upload publicly available PDB files or proprietary structural data files in PDB format. If you want, please check this option and choose your own coordinates for file upload. Such files must be in strict PDB format (see RCSB website for guidance).

  • Option D: Receive the result of predictions by Email


    If you want to receive the results of your predictions by Email, please check this option, and enter your email address correctly.

Example of Server Output

Below is displayed the server prediction results page for the PDB homodimer complex, e.g 1cdt. For each prediction the "Best Patch Cut Off" number was selected as the default value 10, and hence the top 10 scoring patches are displayed..



The example shown is PDB entry ID ; 1cdt (Title CARDIOTOXIN V4/II FROM NAJA MOSSAMBICA MOSSAMBICA). The patch centred on amino acid residue number 45 is selected as the best patch with a combined score 82.75.
For further details of the other residues contained within each patch, please click the icon to "View Patch" , a new window to display an interesting patch on the surface of a protein is opened. In the new window, a Jmol viewer has been implemented that allows the user to view the location of the top scoring patches on the three-dimensional structure of the protein. e.g for patch 1 in this example you will see the following window;



The Jmol molecular graphics viewer version 1.0, which is a java web browser applet that runs on multi-platfom - Windows, Mac, Linux/Unix etc.., is used to display patches. the web browser which supports JAVA such as Internet Explorer(Win32), Mozilla/Firefox(Win32, OSX) ans Safari(OSX) is just required to execute the JmolApplet contained in the web page to view patches. More information, please see the Jmol web site.

Display Types

spacefill (Van der Walls radius)
Spacefill displays the Van der Waals radius of each atom as a solid sphere by default.
spacefill + solvent surface (dots )
Solvent surface displays the solvent accessible surface, which is traced out by the centre of a probe sphere when it is rolled over the surface of a protein.
ribbon + solvent surface
Ribbon displays the polypeptide backbone of a protein as a wide, flattened band.
wireframe + solvent surface
Wirefram display the covalent bonds (except S-S bonds) between all atoms of a molecule.

In addition, SHARP2 server allows users to download the following files;
  1. the file of all best patches information containing residue names and residue numbers.
  2. the file of all parameter values for each patch
  3. Rasmol script files for each patch: These script file colour the 3d structure in Rasmol so that the selected surface patch can be easily viewed.
If you want to download those files, please click icon.

♣ top page
The SHARP2 is operated by Bioinformatics Research Group,
Department of Biochemistry, School of Life Sciences, University of Sussex.
©SHARP2 Any technical problems, please contact: Dr Sue Jones