Automatic generation of gene finders for eukaryotic species
2006
Automatic Gene Finder Generation for Eukaryotic Species
publication
Evidence: moderate
Author Information
Author(s): Munch Kasper, Krogh Anders
Primary Institution: Bioinformatics Centre, University of Copenhagen
Hypothesis
Can we automate the generation of species-specific gene finders using reliable mRNA sequences and genomes?
Conclusion
The Agene method produces gene finders that perform as well or better than manually optimized gene finders.
Supporting Evidence
- Agene automates the creation of gene finders, making it accessible for non-experts.
- The performance of Agene is comparable to existing gene finders.
- The method integrates training set generation and model customization.
Takeaway
This study shows how to create tools that can automatically find genes in new organisms using existing mRNA data, making it easier for scientists.
Methodology
The study uses a Hidden Markov model to create gene predictors from mRNA sequences and genomes.
Limitations
The method may not effectively model gene structures in species with very long introns.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website