RNA 3D Structure Analysis and Prediction

Volume 27 of the series Nucleic Acids and Molecular Biology pp 281-298


Nonredundant 3D Structure Datasets for RNA Knowledge Extraction and Benchmarking

  • Neocles B. LeontisAffiliated withDepartment of Chemistry, Bowling Green State University
  • , Craig L. ZirbelAffiliated withDepartment of Mathematics and Statistics, Bowling Green State University Email author 

* Final gross prices may vary according to local VAT.

Get Access


The continual improvement of methods for RNA 3D structure modeling and prediction requires accurate and statistically meaningful data concerning RNA structure, both for extraction of knowledge and for benchmarking of structure predictions. The source of sufficiently accurate structural data for these purposes is atomic-resolution X-ray structures of RNA nucleotides, oligonucleotides, and biologically functional RNA molecules. All of our basic knowledge of bond lengths, angles, and stereochemistry in RNA nucleotides, as well as their interaction preferences, including all types of base-pairing, base-stacking, and base-backbone interactions, is ultimately extracted from X-ray structures. One key requirement for reference databases intended for knowledge extraction is the nonredundancy of the structures that are included in the analysis, to avoid bias in the deduced frequency parameters. Here, we address this issue and detail how we produce, on a largely automated and ongoing basis, nonredundant lists of atomic-resolution structures at different resolution thresholds for use in knowledge-driven RNA applications. The file collections are available for download at http://​rna.​bgsu.​edu/​nrlist. The primary lists that we provide only include X-ray structures, organized by resolution thresholds, but for completeness, we also provide separate lists that include structures solved by NMR or cryo-EM.