Abstract
BackgroundThe conservation of sequences between related genomes has long been recognised as an indication of functional significance and recognition of sequence homology is one of the principal approaches used in the annotation of newly sequenced genomes. In the context of recent findings that the number non-coding transcripts in higher organisms is likely to be much higher than previously imagined, discrimination between conserved coding and non-coding sequences is a topic of considerable interest. Additionally, it should be considered desirable to discriminate between coding and non-coding conserved sequences without recourse to the use of sequence similarity searches of protein databases as such approaches exclude the identification of novel conserved proteins without characterized homologs and may be influenced by the presence in databases of sequences which are erroneously annotated as coding.ResultsHere we present a machine learning-based approach for the discrimination of conserved coding sequences. Our method calculates various statistics related to the evolutionary dynamics of two aligned sequences. These features are considered by a Support Vector Machine which designates the alignment coding or non-coding with an associated probability score.ConclusionWe show that our approach is both sensitive and accurate with respect to comparable methods and illustrate several situations in which it may be applied, including the identification of conserved coding regions in genome sequences and the discrimination of coding from non-coding cDNA sequences.
Highlights
The conservation of sequences between related genomes has long been recognised as an indication of functional significance and recognition of sequence homology is one of the principal approaches used in the annotation of newly sequenced genomes
Our method calculates various statistics related to the evolutionary dynamics of two aligned sequences
We show that our approach is both sensitive and accurate with respect to comparable methods and illustrate several situations in which it may be applied, including the identification of conserved coding regions in genome sequences and the discrimination of coding from non-coding cDNA sequences
Summary
The conservation of sequences between related genomes has long been recognised as an indication of functional significance and recognition of sequence homology is one of the principal approaches used in the annotation of newly sequenced genomes. While it is probable that representatives of most gene families found in nature have been characterized (at least at the sequence level), lineage specific gene families, genes, and exons - which may be incorporated into messages by alternative splicing and which may not be recovered by ab-initio predictors as components of optimal gene models, are not uncommon (e.g.[4,5,6]). In this context, comparisons between relatively closely related genomes can permit identification of novel exons or coding genes that exhibit low levels of similarity to annotated proteins [7]. Such methods do not rely on the annotation of homologous sequences or the conservation of specific functional signals
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
