Predicting and improving the protein sequence alignment quality by support vector regression
2007

Improving Protein Sequence Alignment with Support Vector Regression

Sample size: 7509 publication Evidence: high

Author Information

Author(s): Lee Minho, Jeong Chan-seok, Kim Dongsup

Primary Institution: Korea Advanced Institute of Science and Technology

Hypothesis

Can support vector regression be used to predict and improve the quality of protein sequence alignments?

Conclusion

The study shows that alignment quality can be predicted accurately, leading to improved protein structure predictions.

Supporting Evidence

  • The Pearson correlation coefficient between observed and predicted MaxSub scores was 0.945.
  • The adaptive selection method improved MaxSub scores by 7.4% compared to using a single best parameter option.
  • The method was tested on a dataset of 7509 protein domains.

Takeaway

This study helps computers figure out the best way to match protein sequences, which is important for understanding how proteins work.

Methodology

The study used support vector regression models to predict alignment quality based on feature vectors derived from protein sequences.

Potential Biases

There is a risk of regression error in selecting alignment options.

Limitations

The method may not cover all possible alignment cases and relies on the MaxSub score, which may not always reflect true alignment quality.

Participant Demographics

The study involved protein sequences from a diverse set of 7509 domains.

Statistical Information

P-Value

p<10-300

Statistical Significance

p<10-300

Digital Object Identifier (DOI)

10.1186/1471-2105-8-471

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication