An introduction to scikit-chem

scikit-chem is a high level cheminformatics library built on rdkit that aims to integrate with the Scientific Python Stack by promoting interoperativity with libraries such as pandas and scikit-learn, and emulating similar patterns and APIs as found in those libraries.

Some notable features include:

  • Pythonic core API

  • Consistent, declarative interfaces for many cheminformatics tasks, including:
    • Reading file formats
    • Chemical standardization
    • Conformer generation
    • Filtering
    • Feature calculation
    • Pipelining
  • A simple interface for chemical datasets

  • Structure visualization

  • Interactivity in Jupyter Notebooks

scikit-chem should be thought of as a simple complement to the excellent rdkit - scikit-chem objects are subclasses of rdkit objects, and as such, the two libraries can usually be used together easily when the advanced functionality of rdkit is required.