Functional Relevance of CASP16 Nucleic Acid Predictions as Evaluated by Structure Providers.

Area of Science:

• Structural Biology
• Computational Biology
• Biophysics

Background:

• Accurate biomolecular structure prediction is vital for understanding function, mutation effects, and drug design.
• Current prediction algorithms struggle with complex biomolecular assemblies involving nucleic acids.

Purpose of the Study:

• To quantitatively and qualitatively evaluate nucleic acid structure predictions for the CASP16 challenge.
• To identify the strengths and limitations of current prediction methods for nucleic acid complexes.

Main Methods:

• Analysis of CASP16 blind prediction targets involving nucleic acids.
• Evaluation by 12 experimental groups who provided the targets.
• Assessment of secondary and tertiary structure prediction accuracy, focusing on functional regions.

Main Results:

• Blind predictions showed accuracy in secondary structures and global folds for some RNAs.
• Predictions lacked accuracy in functionally critical non-canonical regions, including backbone bends and non-standard base pairing.
• Modeling of interfaces between nucleic acids and other molecules (ligands, proteins) was consistently poor.

Conclusions:

• Current prediction algorithms can model basic nucleic acid structures but fail to capture essential functional details.
• Inaccuracies in non-canonical regions and interfaces limit the utility of predictions for understanding RNA function and interactions.
• The dynamic nature of biomolecular complexes presents a future challenge for structure prediction accuracy.