Models and Algorithms for Equilibrium Analysis of Mixed-Material Nucleic Acid Systems

Area of Science:

• Computational Biology
• Biophysics
• Molecular Biology

Background:

• NUPACK software currently analyzes single-material nucleic acid systems (all-RNA or all-DNA).
• Mixed-material nucleic acid systems are vital for advanced in vitro, in situ, and in vivo applications.
• Existing models lack the capability to analyze systems with both RNA and DNA components.

Purpose of the Study:

• To develop physical models and dynamic programming algorithms for analyzing mixed-material nucleic acid systems.
• To enable nucleotide-resolution specification of material composition in complex and test tube ensembles.
• To extend NUPACK's capabilities to RNA/DNA and RNA/2'OMe-RNA systems.

Main Methods:

• Developed new physical models and dynamic programming algorithms for mixed-material systems.
• Constructed free energy parameter sets for RNA/DNA and RNA/2'OMe-RNA by combining existing parameters.
• Implemented new dynamic programming recursions that account for nucleotide-specific material composition.

Main Results:

• Mixed-material algorithms maintain O(N^3) time complexity, similar to single-material algorithms.
• Calculations for complex partition function, equilibrium concentrations, and base-pairing probabilities are now possible for mixed systems.
• Significantly enhanced accuracy in predicting duplex melting temperatures and melt profiles for RNA/DNA and RNA/2'OMe-RNA systems.

Conclusions:

• The new algorithms enable efficient and accurate analysis of mixed-material nucleic acid ensembles.
• These advancements are critical for designing and understanding complex nucleic acid-based applications.
• Mixed-material analyses are accessible via the NUPACK web app and Python module.