Sieving RNA 3D Structures with SHAPE and Evaluating Mechanisms Driving Sequence-Dependent Reactivity Bias

Area of Science:

• Biochemistry
• Structural Biology
• Computational Biology

Background:

• Selective 2'-hydroxyl acylation analyzed by primer extension (SHAPE) is a key technique for RNA flexibility.
• SHAPE is primarily considered a method for RNA secondary structure (2D) determination.
• Existing models show limitations in accurately correlating SHAPE data with native RNA 3D structures.

Purpose of the Study:

• To develop a novel model for improved correlation between SHAPE data and RNA tertiary (3D) structures.
• To enhance the ability to distinguish SHAPE-compatible from incompatible RNA structures.
• To investigate and explain sequence-dependent biases observed in SHAPE experiments.

Main Methods:

• Development of a new computational model relating SHAPE reactivity to RNA 3D structure.
• Validation of the model using RNA decoys to assess structure compatibility.
• Application of replica-exchange umbrella sampling simulations to study SHAPE reagent-RNA interactions.

Main Results:

• The new model demonstrates a significantly higher correlation with native RNA 3D structures compared to previous models.
• The enhanced model improves the discrimination between valid and invalid RNA structures based on SHAPE data.
• A mechanism explaining sequence-dependent bias in SHAPE experiments was identified, linked to reagent binding pocket stability.

Conclusions:

• SHAPE chemical probing contains substantial information about RNA tertiary (3D) structure.
• The developed model offers practical advancements in predicting and understanding RNA 3D conformations from SHAPE data.
• Understanding SHAPE's sequence-dependent biases provides crucial insights for experimental design and data interpretation.