Evaluating Significance Measures for Protein Motifs
Author Information
Author(s): Pedro Gabriel Ferreira, Paulo J. Azevedo
Primary Institution: Department of Informatics, University of Minho, Portugal
Hypothesis
The choice of significance measures affects the evaluation of motifs extracted from protein databases.
Conclusion
This study provides insights on selecting appropriate significance measures for evaluating deterministic motifs in protein databases.
Supporting Evidence
- The study categorizes 14 significance measures for motif evaluation.
- It identifies that some measures are more suitable for highly conserved motifs.
- The research proposes a methodology for combining multiple measures to improve motif ranking.
Takeaway
This study looks at different ways to measure how important patterns in proteins are, helping scientists find the most interesting ones.
Methodology
The study surveys and categorizes 14 significance measures and evaluates their performance in ranking motifs under various conditions.
Potential Biases
Some measures may provide redundant information or be biased towards certain types of motifs.
Limitations
The study may not cover all possible significance measures and their interactions.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website