Predicting Alternative Splicing with ExAlt
Author Information
Author(s): Allen Jonathan E, Salzberg Steven L
Primary Institution: University of Maryland Institute for Advanced Computer Studies
Hypothesis
Can a non-expression based statistical method accurately predict alternatively spliced exons using a single genome sequence and cross-species sequence conservation?
Conclusion
ExAlt successfully identifies the structure of most alternatively spliced exons and shows that cross-species sequence conservation improves prediction precision.
Supporting Evidence
- ExAlt was tested on 1339 Drosophila melanogaster exons from 1160 gene loci.
- 95% of the alternatively spliced test exons were aligned to at least one informant species.
- ExAlt's predictions showed high sensitivity and specificity compared to existing gene finders.
Takeaway
The study created a computer program called ExAlt that helps find parts of genes that can be spliced in different ways, which is important for understanding how genes work.
Methodology
The study used a phylogenetic generalized hidden Markov model to predict alternatively spliced exons based on genomic sequences and cross-species conservation.
Limitations
The method may not accurately predict exons without cross-species conservation evidence, and the performance may vary with different exon types.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website