FRET-guided modeling of nucleic acids

Area of Science:

• Biophysics
• Computational Biology
• Structural Biology

Background:

• RNA functional diversity arises from conformational heterogeneity.
• Mapping structural transitions in nucleic acid ensembles is crucial for understanding function.
• Single-molecule spectroscopy and computational modeling offer complementary approaches.

Purpose of the Study:

• To develop a framework harmonizing single-molecule Förster resonance energy transfer (FRET) measurements with computational modeling.
• To utilize FRET data for filtering de novo RNA structure prediction ensembles.
• To improve the mechanistic understanding of nucleic acid structural dynamics and interactions.

Main Methods:

• Integration of single-molecule FRET experiments with molecular dynamics simulations and de novo structure prediction (Rosetta).
• In silico recreation of FRET experiments using all-atom or implicit fluorophore modeling.
• Application of accessible-contact volumes as a post hoc scoring method for structure prediction.

Main Results:

• Demonstrated FRET's utility in filtering RNA structure prediction ensembles by refuting incompatible models.
• Successfully benchmarked the FRET-assisted modeling approach on DNA and validated it on a dynamic riboswitch.
• Recapitulated the global fold of a riboswitch using a FRET coordinate for four-way junction assembly.

Conclusions:

• Computational fluorescence spectroscopy enhances the interpretability of dynamic structural ensembles.
• The developed pipeline improves mechanistic understanding of nucleic acid interactions.
• This integrated approach provides a powerful tool for studying RNA structure and dynamics.