FebRNA

FebRNA predicts RNA three-dimensional (3D) structures using a fragment-ensemble-based approach to assemble coarse-grained models and reconstruct all-atom RNA structures for modeling diverse RNA topologies.

Key Features:

• Fragment Ensemble Library: Builds a comprehensive library of non-redundant, sequence-independent coarse-grained (CG) fragment ensembles for diverse RNA conformations.
• Coarse-Grained 3D Structure Assembly: Assembles potential CG 3D structures by combining fragment ensembles to explore multiple structural configurations.
• Scoring and Selection: Evaluates assembled CG structures with a CG scoring function to identify top-scored candidate models.
• All-Atom Structure Reconstruction: Reconstructs detailed all-atom 3D models from selected top-scored CG structures.

Scientific Applications:

• Complex RNA topologies: Predicts 3D structures for RNAs with pseudoknots, three-way, four-way, and five-way junctions.
• Benchmarking and blind challenges: Addresses challenging cases from the RNA-Puzzles competition.
• Structure–function analysis: Provides all-atom models to support studies of RNA structure–function relationships.
• Support for related fields: Facilitates research in molecular biology, genetics, and drug design by supplying predicted RNA 3D structures.

Methodology:

Establish non-redundant sequence-independent coarse-grained fragment ensembles, assemble CG 3D structures via fragment assembly, score assembled CG models with a CG scoring function to select top-scored structures, and reconstruct all-atom 3D models from those selected CG structures.