FunFOLD: A New Method for Predicting Protein Binding Sites
Author Information
Author(s): Roche Daniel B, Tetchner Stuart J, McGuffin Liam J
Primary Institution: School of Biological Sciences, University of Reading
Hypothesis
Can an automated method improve the prediction of ligand binding residues in proteins using 3D models?
Conclusion
The FunFOLD software provides competitive predictions for ligand binding site residues and is freely available for both expert and non-expert users.
Supporting Evidence
- The FunFOLD method outperformed all other methods tested at CASP8.
- It is competitive with top server methods at CASP9.
- The software is available as a standalone package and a web server.
- FunFOLD uses a novel automated method for ligand clustering and identification.
Takeaway
FunFOLD is a computer program that helps scientists figure out where proteins bind to other molecules, making it easier to understand how proteins work.
Methodology
The FunFOLD method uses a novel automated approach for cluster identification and residue selection based on 3D models and templates.
Potential Biases
Potential over-prediction of residues due to the voting system used for ligand binding predictions.
Limitations
The method may struggle with proteins that have multiple binding sites or are in disordered regions.
Statistical Information
P-Value
p<0.05
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website