ASH Structure Alignment Package: Sensitivity and Selectivity in Domain Classification
Author Information
Author(s): Daron M. Standley, Hiroyuki Toh, Haruki Nakamura
Primary Institution: Institute for Protein Research, Osaka University
Hypothesis
How can structural differences and similarities be quantified to measure evolutionary relationships between proteins?
Conclusion
The ASH structure alignment package shows high selectivity and sensitivity for domain classification, making it a practical option for large-scale structure classification studies.
Supporting Evidence
- The ASH score achieved an area of .96 under the ROC curve.
- The ASH program is competitive in CPU cost with the fastest programs.
- The study identified weak sequence homologies that were undetectable by traditional methods.
Takeaway
The ASH program helps scientists compare protein structures to find out how they are related, even if they look different.
Methodology
The study developed a scoring function using a training set of sequence-unique CATH and SCOP domains to improve sensitivity and selectivity in protein structure alignment.
Potential Biases
There may be risks of bias in the classification of borderline cases where alignment scores do not clearly indicate true relationships.
Limitations
The study may have limitations related to the binary classification of domains and the potential for overfitting the training set.
Statistical Information
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website