A framework for protein structure classification and identification of novel protein structures
2006

A Framework for Protein Structure Classification

Sample size: 5289 publication Evidence: high

Author Information

Author(s): Kim You Jung, Patel Jignesh M

Primary Institution: University of Michigan

Hypothesis

Can an automated method accurately classify and identify novel protein structures?

Conclusion

The proCC method effectively classifies new protein domains and suggests new domain families with high accuracy.

Supporting Evidence

  • The proCC method achieved 86.0%, 87.7%, and 90.5% classification accuracy at family, superfamily, and fold levels respectively.
  • The method is about 15-19% more accurate than existing methods like SGM and comparable to SCOPmap.
  • It produced clusters that closely correspond to new families in SCOP 1.69.

Takeaway

This study created a computer program that helps scientists sort and understand new proteins by comparing their shapes to known ones.

Methodology

The study used a three-component framework involving structure comparison, classification, and clustering to classify protein domains.

Limitations

The method cannot completely replace manual classification and may not capture all nuances of protein classification.

Digital Object Identifier (DOI)

10.1186/1471-2105-7-456

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication