CATHEDRAL: A fast and effective algorithm to predict folds and domain boundaries from multidomain protein structures
2007

CATHEDRAL: A Fast and Effective Algorithm to Predict Folds and Domain Boundaries from Multidomain Protein Structures

Sample size: 964 publication 10 minutes Evidence: high

Author Information

Author(s): Redfern Oliver C, Harrison Andrew, Dallman Tim, Pearl Frances M. G, Orengo Christine A

Primary Institution: University College London

Hypothesis

CATHEDRAL aims to improve the identification of protein folds and domain boundaries in multidomain protein structures.

Conclusion

CATHEDRAL successfully identifies known folds in novel multidomain structures with high accuracy and low false positive rates.

Supporting Evidence

  • CATHEDRAL locates known folds in 90% of cases with less than 1% false positives.
  • 80% of assigned domains have boundaries correctly delineated within ten residues.
  • CATHEDRAL outperforms traditional methods like hidden Markov models in domain detection.

Takeaway

CATHEDRAL is a computer program that helps scientists find parts of proteins that fold in specific ways, making it easier to understand how proteins work.

Methodology

CATHEDRAL uses a combination of graph theory and dynamic programming to identify and align protein domains against a library of known folds.

Potential Biases

Potential bias may arise from the reliance on existing fold classifications and the iterative nature of the algorithm.

Limitations

CATHEDRAL may miss some domains, particularly small or distorted structures, and relies on the quality of the CATH database.

Statistical Information

P-Value

p<0.05

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1371/journal.pcbi.0030232

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication