Characterization and visualization of RNA secondary structure Boltzmann ensemble via information theory

• DOI: 10.1186/s12859-018-2078-5
• Publication Date: 2018-03-05T06:26:30Z
• Authors: Luan Lin; Wilson H. McKerrow; Bryce Richards; Chukiat Phonsom; Charles E. Lawrence
• Keywords: Characterization; Nucleic acid structure

The nearest neighbor model and associated dynamic programming algorithms allow for the efficient estimation of RNA secondary structure Boltzmann ensemble. However because a given only contains fraction possible helices that could form from sequence, ensemble is multimodal. Several methods exist clustering structures finding those modes. less focus to exploring underlying reasons this multimodality: presence conflicting basepairs. Information theory, or more specifically mutual information, provides method identify basepairs are key structure. To end we find most informative visualize effect these on Knowing whether basepair present tells us not status particular pair but also large amount information about which other pairs present. We few account structural uncertainty. identification indicates small changes sequence stability will have provide novel algorithm uses lead multimodal distribution. then overall