Related Experiment Videos
Frequency of gaps observed in a structurally aligned protein pair database suggests a simple gap penalty function
Nalin C W Goonesekere1, Byungkook Lee
1Laboratory of Molecular Biology, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Building 37, Room 5120, 37 Convent Drive MSC 4264, Bethesda, MD 20892-4264, USA.
Nucleic Acids Research
|May 25, 2004
Summary
Analyzing protein sequence alignment, this study reveals a bilinear pattern in gap frequencies, suggesting a revised affine gap penalty. This finding improves homology searches and protein alignment accuracy.
Area of Science:
- Bioinformatics
- Computational Biology
- Structural Biology
Background:
- Accurate protein sequence alignment is crucial for homology searches.
- Current methods use heuristic affine gap penalties (q + r*n).
- A more rational scoring scheme is needed for improved accuracy.
Purpose of the Study:
- To investigate the natural distribution of gaps in protein sequences.
- To develop a more biologically informed gap penalty function.
- To enhance the accuracy of protein homology searches and alignments.
Main Methods:
- Analysis of a database of structurally aligned protein domain pairs.
- Examination of gap frequency distribution in relation to gap length.
- Application of linear regression to model gap frequency patterns.
- Categorization of gaps based on flanking secondary structures.
Main Results:
- Gap frequency logarithm shows a bilinear relationship with gap length, with a breakpoint at length 3.
- Two linear regression lines (R² values of 1.0 and 0.99) accurately approximate this pattern.
- This bilinear behavior is consistent across different secondary structure contexts and independent datasets.
Conclusions:
- The observed bilinear gap distribution suggests limitations in the current affine gap penalty model.
- A modified gap penalty function based on this bilinear behavior could improve sequence alignment algorithms.
- Findings support a more data-driven approach to developing scoring schemes in bioinformatics.