BiORSEO: a bi-objective method to predict RNA secondary structures with pseudoknots using RNA 3D modules

Area of Science:

• Computational Biology
• Bioinformatics
• Structural Biology

Background:

• RNA secondary structure prediction is crucial for understanding RNA function.
• Existing methods often rely solely on energy-based models.
• RNA 3D structures reveal recurrent interaction patterns known as RNA modules.

Purpose of the Study:

• To explore the utility of RNA modules for enhancing RNA secondary structure prediction.
• To develop a novel bi-objective prediction method integrating energy and knowledge-based potentials.

Main Methods:

• Proposed a bi-objective optimization approach (BiORSEO) minimizing energy and knowledge-based potentials.
• Integrated information from RNA modules sourced from Rna3Dmotif and RNA 3D Motif Atlas.
• Evaluated different scoring strategies for module insertions (size, complexity, probability).

Main Results:

• Benchmarked BiORSEO against state-of-the-art tools using known secondary structures.
• Demonstrated the effectiveness of incorporating RNA modules into secondary structure prediction.
• Highlighted the utility of a hybrid physics-based and data-driven approach.

Conclusions:

• The BiORSEO tool offers improved RNA secondary structure prediction by integrating RNA modules.
• The developed method provides optimal solutions from a Pareto set, balancing energy and knowledge-based information.
• Software and datasets are publicly available for further research and application.