Secondary structural entropy in RNA switch (Riboswitch) identification.

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• RNA regulatory elements, particularly riboswitches, are crucial for gene regulation in bacteria.
• Riboswitches control gene expression through ligand-induced conformational changes.
• Current methods for riboswitch identification rely on sequence and structural homology, limiting discovery of novel or diverse elements.

Purpose of the Study:

• To explore structural entropy as a novel feature for identifying riboswitches.
• To develop and test computational methods for riboswitch discovery based on structural entropy.

Main Methods:

• Structural entropy of riboswitch sequences was calculated to assess secondary structural dynamics.
• Entropy values were compared against mutants, dinucleotide shuffles, and reverse complements using stochastic context-free grammar folding models.
• Classifiers integrating structural entropy with sequence and structural features were developed and validated on bacterial genomes (Bacillus subtilis, Escherichia coli, Synechococcus elongatus).

Main Results:

• A relationship was observed between higher structural entropy and the propensity of RNA sequences to adopt alternative structures.
• Genome-wide analyses in Bacillus subtilis and Escherichia coli identified significant structural entropy values in specific genomic regions, including untranslated regions of genes like cotH and sucC.
• The structural entropy approach demonstrated potential as a riboswitch identifier, complementing existing methods.

Conclusions:

• Structural entropy serves as a viable, albeit modest, indicator of RNA sequence potential for alternative structures.
• This finding supports the utility of structural entropy in exploring RNA conformational dynamics.
• Further research is needed to fully understand structural entropy across diverse RNA sequences and folding models for enhanced riboswitch discovery.