PDB-UF: database of predicted enzymatic functions for unannotated protein structures from structural genomics
2006

PDB-UF: Database of Predicted Enzymatic Functions for Unannotated Protein Structures

Sample size: 499 publication Evidence: moderate

Author Information

Author(s): Marcin von Grotthuss, Dariusz Plewczynski, Krzysztof Ginalski, Leszek Rychlewski, Eugene I Shakhnovich

Primary Institution: Harvard University

Hypothesis

Can we predict the enzymatic functions of unannotated protein structures using structural similarity?

Conclusion

The PDB-UF database will help reduce the number of protein structures with unknown functions by providing structure-based predictions of enzymatic properties.

Supporting Evidence

  • The PDB-UF database is a web-accessible collection of predictions for enzymatic properties.
  • The study suggests using different similarity score cutoffs for different protein folds.
  • The algorithms used can reduce the rate of false positive assignments.

Takeaway

Scientists created a database to guess the jobs of proteins that we don't know much about by looking at their shapes.

Methodology

The study used two algorithms, 3D-Hit and 3D-Fun, to assign EC numbers to protein structures based on structural similarity.

Potential Biases

Conflicting predictions between the two algorithms may lead to inconsistencies in function assignments.

Limitations

The predictions may not be entirely applicable to all protein structures due to the unique nature of the targeted structures in structural genomics.

Digital Object Identifier (DOI)

10.1186/1471-2105-7-53

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication