Improving Protein Sequence Alignment with Correlated Mutation
Author Information
Author(s): Jeong Chan-seok, Kim Dongsup
Primary Institution: KAIST
Hypothesis
Can incorporating correlated mutation information improve the quality of protein sequence alignment?
Conclusion
The CM profile method significantly enhances alignment quality, especially for distantly related proteins with low sequence identity.
Supporting Evidence
- Combining CM profile with sequence profile improves alignment quality by 13.9%.
- Using CM profile alone performs poorly, but it significantly enhances alignment when combined with other methods.
- CM profile is particularly effective for proteins with low sequence identity.
Takeaway
This study shows that using information about how mutations in proteins are related can help us align them better, especially when they are not very similar.
Methodology
The study developed a method called CM profile that uses linear predictive coding to represent correlated mutations and combines it with conventional sequence profiles for better alignment.
Limitations
The method's performance may decrease with fewer sequences in multiple sequence alignments.
Statistical Information
P-Value
2.1e-252
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website