Structural constraints identified with covariation analysis in ribosomal RNA

Area of Science:

• Computational Biology
• Bioinformatics
• RNA Structure Prediction

Background:

• Covariation analysis in RNA sequences identifies positional constraints, often linked to base pairs in helical structures.
• Mutual information (MI) is effective for predicting RNA secondary structure and some tertiary interactions.
• Phylogenetic event counting methods offer greater sensitivity and confidence in base pair prediction compared to early MI approaches.

Purpose of the Study:

• To develop and apply a novel phylogenetic event counting (PEC) method for quantifying positional covariation in RNA sequences.
• To integrate the PEC method with the Gutell lab's RNA Comparative Analysis Database (rCAD).
• To compare the performance of PEC and MI methods in identifying RNA base pairs and explore broader structural constraints.

Main Methods:

• Development of a novel phylogenetic event counting (PEC) method for RNA sequence analysis.
• Application of PEC and mutual information (MI) methods to identify positional covariation.
• Utilized an N-best and helix-extension strategy in conjunction with both methods to maximize base pair identification.

Main Results:

• The PEC method, when used with rCAD, effectively quantifies positional covariation.
• PEC and MI methods uniquely identify different base pairs, and their combined use maximizes the number of predicted base pairs.
• Covariation analysis reveals significant, albeit weaker, correlations between nucleotides in proximity in 3D RNA structures, indicating constraints beyond base pairs.

Conclusions:

• The novel PEC method provides a powerful tool for RNA structure prediction and analysis.
• Combining PEC and MI methods with advanced strategies maximizes the identification of RNA base pairs.
• Covariation analysis extends beyond predicting base pairs to identify other structural constraints in three-dimensional RNA folding.