New Algorithm for Sequence Alignment from Superimposed Structures
Author Information
Author(s): Tai Chin-Hsien, Vincent James J, Kim Changhoon, Lee Byungkook
Primary Institution: National Cancer Institute, National Institutes of Health
Hypothesis
Can the Seed Extension algorithm produce more accurate sequence alignments from superimposed structures compared to traditional dynamic programming methods?
Conclusion
The Seed Extension algorithm is fast and produces more accurate sequence alignments from superimposed structures than three other programs tested that use dynamic programming algorithms.
Supporting Evidence
- The SE algorithm achieved an average accuracy of 95.9% over 582 pairs of superimposed proteins.
- SE produced alignments up to 18% more accurate than the next best scoring program for low similarity pairs.
- When implemented in SHEBA, SE improved alignment accuracy by 10% on average for certain structure pairs.
Takeaway
This study created a new way to match sequences from two structures that are similar, and it works better and faster than older methods.
Methodology
The Seed Extension algorithm identifies 'seeds' of structurally equivalent residues and extends these to create alignments without using gap penalties.
Limitations
The algorithm's performance may vary with different databases and may require parameter adjustments for optimal results.
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