Computational identification of noncoding RNAs in E. coli by comparative genomics

Area of Science:

• Genomics
• Bioinformatics
• Molecular Biology

Background:

• Noncoding RNA (ncRNA) genes are crucial for cellular functions but are difficult to detect using standard genomic analysis.
• Unlike protein-coding genes, ncRNA genes lack obvious statistical signatures, making them 'invisible' to traditional gene-finding methods.

Purpose of the Study:

• To develop and apply a computational comparative genomic strategy for identifying novel noncoding RNA genes.
• To overcome the limitations of detecting ncRNA genes in genome sequences.

Main Methods:

• Developed a computational screen utilizing probabilistic models to identify conserved RNA secondary structures.
• Applied the method to pairwise sequence alignments of intergenic regions in four related bacteria, including Escherichia coli.
• Distinguished conserved RNA structures from other conserved sequences based on expected mutational patterns.

Main Results:

• Screened over 23,000 conserved interspecies pairwise alignments, predicting 275 candidate structural RNA loci.
• Experimentally assayed 49 candidate loci, confirming expression of small, noncoding RNA transcripts from at least 11 loci.
• Demonstrated the feasibility of a whole-genome screen for ncRNA genes.

Conclusions:

• The developed computational approach is effective for discovering structural noncoding RNA genes.
• This method can be applied to any genome with available comparative sequence data.
• Identified novel ncRNA transcripts in Escherichia coli, expanding the known repertoire of functional RNAs.