DETECT--a density estimation tool for enzyme classification and its application to Plasmodium falciparum
Stacy S Hung1, James Wasmuth, Christopher Sanford
1Program in Molecular Structure and Function, Hospital for Sick Children, 15-704 MaRS TMDT East, 101 College Street, Toronto, ON M5G 1L7, Canada.
Bioinformatics (Oxford, England)
|June 2, 2010
Summary
DETECT improves enzyme prediction accuracy using a probabilistic method that considers sequence diversity. This approach enhances gene annotation reliability and identifies novel enzymes, particularly in organisms like Plasmodium falciparum.
Area of Science:
- Genomics
- Bioinformatics
- Enzymology
Background:
- Accurate gene annotation in genomics is challenging.
- Current homology-based enzyme prediction methods are error-prone and lack reliability measures.
Purpose of the Study:
- To present DETECT, a novel probabilistic method for enzyme prediction.
- To improve the accuracy and reliability of enzyme annotation.
- To identify potential annotation errors and novel enzymes.
Main Methods:
- DETECT accounts for sequence diversity across enzyme families.
- Compares global alignment scores of unknown proteins to known enzymes.
- Calculates an integrated likelihood score to rank relevant reaction classes.
Main Results:
- DETECT demonstrates significant improvements in enzyme annotation accuracy compared to BLAST.
- The method successfully identifies potential annotation errors.
- Novel enzymes of therapeutic interest were predicted in Plasmodium falciparum.
Conclusions:
- DETECT offers a more reliable and accurate approach to enzyme prediction.
- The method has implications for understanding gene function and identifying drug targets.
- DETECT is available as a standalone application for broader use.
