Large scale clustering of protein sequences with FORCE -A layout based heuristic for weighted cluster editing
2007
FORCE: A New Method for Clustering Protein Sequences
Sample size: 192187
publication
Evidence: high
Author Information
Author(s): Wittkop Tobias, Baumbach Jan, Lobo Francisco P, Rahmann Sven
Primary Institution: Bielefeld University
Hypothesis
The model of weighted cluster editing is well-suited to protein clustering.
Conclusion
FORCE is an effective alternative to existing clustering algorithms and can handle large datasets efficiently.
Supporting Evidence
- FORCE outperforms popular clustering tools like Spectral clustering and TribeMCL.
- FORCE can handle 192,187 prokaryotic protein sequences from the COG database.
- FORCE is open source and implemented in Java.
Takeaway
The FORCE method helps group similar proteins together based on their sequences, making it easier to understand their functions.
Methodology
The FORCE heuristic uses a graph-based approach to cluster protein sequences based on pairwise similarity measures.
Limitations
The method uses a single threshold for all clusters, which may not be optimal for all cases.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website