Guiding Similarity Search in Chemical Fragment Spaces with Weighted Fingerprints

Area of Science:

• Computational chemistry
• Cheminformatics
• Drug discovery

Background:

• Chemical fragment spaces enable larger compound libraries.
• The SpaceLight algorithm introduced efficient similarity search in these spaces.
• Large chemical spaces necessitate advanced search customization.

Purpose of the Study:

• Introduce weighted SpaceLight, an enhancement for focused similarity searches.
• Improve structural control over search results in chemical fragment spaces.
• Provide a method to avoid postfiltering millions of hits.

Main Methods:

• Enhancement of the SpaceLight algorithm to incorporate weighting.
• Focusing search on important areas of a query molecule.
• Adapting the algorithm for diverse search scenarios.

Main Results:

• Weighted SpaceLight increases local similarity while allowing variability.
• The method produces more molecules preserving selected substructures.
• Demonstrates adaptability for different search requirements.

Conclusions:

• Weighted SpaceLight offers enhanced control for chemical space exploration.
• Bridges the gap between global similarity and substructure focus.
• Provides a valuable tool for drug discovery research.