Related Experiment Videos
Systematic and fully automated identification of protein sequence patterns.
R K Hart1, A K Royyuru, G Stolovitzky
1IBM Computational Biology Center, T.J. Watson Research Center, Yorktown Heights, NY 10598, USA. reece@in-machina.com
Summary
An efficient algorithm automatically identifies protein sequence patterns using the Splash algorithm and statistical significance testing. This method aids in curating motif and profile databases, improving pattern discovery and biological relevance assessment.
Area of Science:
- Bioinformatics
- Computational Biology
- Genomics
Background:
- Protein sequence families are crucial for understanding protein function and evolution.
- Existing methods for pattern discovery in protein families can be limited in efficiency and accuracy.
- Automated identification of statistically significant patterns is essential for database curation and biological insight.
Purpose of the Study:
- To present an efficient, automated algorithm for discovering patterns in protein sequence families.
- To assess the statistical significance of discovered patterns.
- To demonstrate the algorithm's utility in curating protein motif and profile databases.
Main Methods:
- Utilized the Splash deterministic pattern discovery algorithm.
- Developed a framework for assessing the statistical significance of identified patterns.
- Applied the algorithm to 974 PROSITE families.
Main Results:
- Splash algorithm improved pattern specificity or sensitivity in 28% of tested PROSITE families.
- In 75% of cases, Splash patterns significantly overlapped with existing PROSITE patterns.
- Statistical significance of patterns correlated well with their biological significance, as shown in serine proteases.
Conclusions:
- The developed algorithm offers an efficient and automated approach to protein pattern discovery.
- The method is suitable for daily curation of motif and profile databases.
- Discovered patterns have strong statistical and biological relevance, with applications in multiple sequence alignment and motif model training.