A framework and predictor based on support vector machine and random walk graph kernel for scoring protein-protein interfaces.

235 commits | Last update: June 23, 2020

Cite this software

Choose a version:
[[ releases.length > 0 ? releases[selectedIndex].doi : conceptDOI ]]
Copy to clipboard
Choose a reference manager file format:
Download file

What iScore can do for you

  • Provides a state-of-the-art scoring predictor trained on the data of Docking Benchmark 5 (
  • Provides easy-to-use interface to train a new scoring predictor
  • Easily extended with more different and various features of protein interface
  • Active and competitive player in the CAPRI challenge for 3D structure predictions of protein complexes (

iScore offers simple solutions to classify protein-protein interfaces using a support vector machine approach on graph kernels. The simplest way to use iScore is through dedicated binaries that hide the complexity of the approach and allows access to the code with simple command line interfaces. The two binaries are iscore.train and iscore.predict that respectively train a model using a trainging set and use this model to predict the near-native character of unkown conformations.

Read more
  • Machine learning
  • Big data
Programming Language
  • Python
  • Apache-2.0
Source code

Participating organizations


  • Cunliang Geng
    Netherlands eScience Center
  • Nicolas Renaud
    Netherlands eScience Center
  • Yong Jung
  • Li Xue
    Utrecht University
Contact person
Nicolas Renaud
Netherlands eScience Center

Information for page maintainers

OAI-PMH metadata:
429 Client Error: TOO MANY REQUESTS for url:
citation metadata: