metaPPI is a web server for protein-protein interaction binding site prediction!

Input methods: Help
  1. First please enter your e-mail address
  2. Sumbit the protein structure: upload the pdbfile or enter the PDB ID
  3. Specify the chain ID, only one chain ID is allowed
  4. Click on the "Predict" button
  5. The server will return you the prediction results as B factor values in the returned PDB file
Enter your e-mail address: Help
     

Submit your protein structure: Help
a. upload a PDB file:
     
b. or enter a PDB ID:
        (For example 1acb)

Specify the chain ID: Help
     

  
Note:
It might take up to 10 minutes to predict the binding sites depending on the size of your protein and the status of other predict servers. Please be patient.
The algorithm of metaPPI is illustrated here.

Other prediction servers:
We use the prediction results from the following prediction servers to build up our meta prediction method.
  • PINUP, empirical scoring function
  • PPI-PRED, SVM-based
  • cons-PPISP, neural network-based
  • Promate, empirical scoring function
  • SPPIDER, SVM, neural network
    We appreciate the authors of the above servers.
    If you have your own prediction server and want to be integrated into metaPPI, please contace me.

    Reference:
    If you find this server useful for your work, please cite: Bingding Huang and Michael Schroeder(2008), Using protein binding site to improve protein-protein docking Gene, Epub 1;422(1-2):14-21. link
    Contact:
    Bingding Huang, e-mail: bhuang@biotec.tu-dresden.de, homepage.

    You might also be interested in another server we developed for protein ligand binding site prediction: LIGSITEcsc.

    you are the 5872th visitor since 1th, June, 2007.

    Last update: 5th, June, 2007


    Bioinformatics group                 Biotec             TU Dresden
    Bioinformatics group                     Biotec                                 TU Dresden