Improving Protein Domain Assignments
Author Information
Author(s): Jessica H. Fong, Aron Marchler-Bauer
Primary Institution: National Center for Biotechnology Information, National Library of Medicine, National Institutes of Health
Hypothesis
Can a new method for assigning protein domains reduce misclassifications in protein sequence databases?
Conclusion
The proposed domain subfamily assignment rule has significantly improved the accuracy of domain annotations in protein sequences.
Supporting Evidence
- The proposed method achieved 96% accuracy in domain assignments.
- Using score thresholds reduced misclassification rates significantly.
- The method has been incorporated into the CD-Search software for improved protein annotation.
Takeaway
This study helps scientists better classify proteins by using a new method that reduces mistakes in labeling protein functions.
Methodology
The study analyzed alignment scores from NCBI-curated domain assignments and proposed heuristics for better classification.
Potential Biases
Potential bias from longer profiles and missing subfamilies could affect classification accuracy.
Limitations
The study may not account for all possible domain subfamilies due to incomplete databases.
Statistical Information
P-Value
0.85
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website