Abstract
This paper introduces a new computer system for recognition of functional transcription start sites (TSSs) in RNA polymerase II promoter regions of vertebrates. This system allows scanning complete vertebrate genomes for promoters with significantly reduced number of false positive predictions. It can be used in the context of gene finding through its recognition of the 5′ end of genes. The implemented recognition model uses a composite-hierarchical approach, artificial intelligence, statistics, and signal processing techniques. It also exploits the separation of promoter sequences into those that are C+G-rich or C+G-poor. The system was evaluated on a large and diverse human sequence-set and exhibited several times higher accuracy than several publicly available TSS-finding programs. Results obtained using human chromosome 22 data showed even greater specificity than the evaluation set results. The system has been implemented in the Dragon Promoter Finder package, which can be accessed at http://sdmc.krdl.org.sg:8080/promoter/.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 323-332 |
| Number of pages | 10 |
| Journal | Journal of Molecular Graphics and Modelling |
| Volume | 21 |
| Issue number | 5 |
| DOIs | |
| State | Published - Mar 2003 |
| Externally published | Yes |
Keywords
- Eukaryotic promoters
- Promoter modelling
- Promoter recognition
- Transcription start site
ASJC Scopus subject areas
- Spectroscopy
- Physical and Theoretical Chemistry
- Computer Graphics and Computer-Aided Design
- Materials Chemistry