GARN2

GARN2 predicts three-dimensional structures of large RNA molecules at coarse-grained resolution to enable analysis of RNA folding and structure–function relationships.

Key Features:

• Coarse-Grained Modeling: Employs a coarse-grained representation of RNA to retain essential structural features while reducing computational complexity.
• Input Flexibility: Accepts secondary structure information as input to generate 3D models from known or predicted secondary structures.
• Structure Sampling and Refinement: Implements a two-step process that samples possible 3D structures and then extracts the two candidate structures closest to the native conformation.
• Regret Minimization Optimization: Leverages regret minimization techniques to enhance prediction accuracy.
• Performance Improvements: Offers improved speed and accuracy relative to its predecessor and other state-of-the-art methods.
• Scalability: Designed to be efficient and scalable for large RNA molecules.

Scientific Applications:

• Structural Biology: Provides coarse-grained 3D models to support analyses of RNA folding and tertiary architecture.
• Molecular Genetics: Aids investigation of RNA-mediated regulatory mechanisms through structural hypotheses.
• Biochemistry: Supplies structural models for studying RNA interactions and mechanistic proposals.
• RNA Therapeutics and Diagnostics Design: Offers preliminary 3D conformations to inform design and evaluation of RNA-based therapeutics and diagnostics.

Methodology:

Represents RNA at coarse-grained resolution, samples ensembles of 3D structures, selects the two candidate structures closest to the native conformation from the sampled set, and applies regret minimization techniques to improve prediction accuracy.