Effectiveness of 2D fingerprints for scaffold hopping
Eleanor J Gardiner1, John D Holliday, Caroline O'Dowd
1Information School, University of Sheffield,Western Bank, Sheffield, S10 2TN, UK.
Future Medicinal Chemistry
|April 2, 2011
Summary
Two-dimensional (2D) fingerprints are effective for scaffold hopping, enabling the identification of novel chemical structures. The extended connectivity fingerprint (ECFP4) often yields the best results in drug discovery programs.
Area of Science:
- Medicinal Chemistry
- Computational Chemistry
- Drug Discovery
Background:
- Scaffold hopping is a key strategy in drug discovery for identifying novel chemical entities.
- Traditional similarity searching using 2D fingerprints has been questioned for its suitability in scaffold hopping.
Purpose of the Study:
- To evaluate the effectiveness of six common 2D fingerprints for scaffold-hopping similarity searches.
- To assess the ability of 2D fingerprints to identify novel scaffolds in diverse chemical databases.
Main Methods:
- Systematic evaluation of six distinct 2D fingerprint types.
- Application of similarity searches across three large chemical databases: MDL Drug Data Report, World of Molecular Bioactivity, and Maximum Unbiased Validation.
- Analysis of scaffold hopping performance and identification of novel chemotypes.
Main Results:
- 2D fingerprints demonstrated significant utility in scaffold hopping, successfully identifying novel scaffolds in most searches.
- The enrichment of novel scaffolds was influenced by the structural diversity of the target active compounds.
- The extended connectivity fingerprint encoding a circular substructure of diameter four bonds (ECFP4) frequently provided the highest enrichment.
Conclusions:
- 2D fingerprints offer a computationally efficient and straightforward method for identifying novel chemotypes.
- These findings support the use of 2D fingerprints in lead discovery programs for exploring diverse chemical spaces and achieving scaffold hopping.
