Data Structures for Genome Annotation, Alternative Splicing, and Validation
To establish a clean basis for studying alternative splicing and gene regulation in life science projects, a powerful data modeling and also a strict validation procedure for assigning levels of reliability to given gene models is essential. One common problem of public genome databases are insuffic...
Gespeichert in:
Hauptverfasser: | , , |
---|---|
Format: | Buchkapitel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | To establish a clean basis for studying alternative splicing and gene regulation in life science projects, a powerful data modeling and also a strict validation procedure for assigning levels of reliability to given gene models is essential. One common problem of public genome databases are insufficiently organized and linked description data, which make it difficult to study relations of the alternative isoforms of a gene that are relevant for medi cine and plant genome research. This is a severe obstacle for the integration of biological data and motivated us to establish a new modeling instance and that we call splice template or sTMP. Every sTMP has a unique splicing pattern, but the length of the first and the last exon remains undefined. This allows to model different gene isoforms with the same splicing pattern. By utilizing this more fine-grained data structure, many cases of plurivalent mRNA-CDS relations are uncovered. There are more than 3,000 extra CDSs in the human genome compatible with the categories sTMP, mRNA and CDS, which exceed the classical one-to-one relations of mRNAs and CDSs. In one case, 11 extra CDSs are compatible with one mRNA. Crosslinks between mRNAs derived from different sTMPs leading to the same CDS are now accessible as well as disease-related ruptures in UTR regions. This allows discovering and validating disease and tissue specific differences in alternative splicing, gene expression and regulation. Another problem in public databases is a too much relaxed standard for labeling genes “confirmed by ESTs and full-length-cDNAs.” We provide a pipeline that handles gene annotations from different sources, integrates them into complex gene models and assigns strict validation tags, constrained by a local low-error model for the alignments of genome annotation and transcripts. The data structures are being implemented and made publicly available at the Plant Data Warehouse of the Bioinformatics Center Gatersleben-Halle (http://portal.bic-gh.de/sTMP). |
---|---|
ISSN: | 0302-9743 1611-3349 |
DOI: | 10.1007/11799511_11 |