Computational prediction of riboswitch tertiary structures including pseudoknots by RAGTOP: A hierarchical graph sampling approach

Namhee Kim, Mai Zahran, Tamar Schlick

Research output: Contribution to journalArticle

Abstract

The modular organization of RNA structure has been exploited in various computational and theoretical approaches to identify RNA tertiary (3D) motifs and assemble RNA structures. Riboswitches exemplify this modularity in terms of both structural and functional adaptability of RNA components. Here, we extend our computational approach based on tree graph sampling to the prediction of riboswitch topologies by defining additional edges to mimick pseudoknots. Starting from a secondary (2D) structure, we construct an initial graph deduced from predicted junction topologies by our data-mining algorithm RNAJAG trained on known RNAs; we sample these graphs in 3D space guided by knowledge-based statistical potentials derived from bending and torsion measures of internal loops as well as radii of gyration for known RNAs. We present graph sampling results for 10 representative riboswitches, 6 of them with pseudoknots, and compare our predictions to solved structures based on global and local RMSD measures. Our results indicate that the helical arrangements in riboswitches can be approximated using our combination of modified 3D tree graph representations for pseudoknots, junction prediction, graph moves, and scoring functions. Future challenges in the field of riboswitch prediction and design are also discussed.

Original languageEnglish (US)
Pages (from-to)115-135
Number of pages21
JournalMethods in Enzymology
Volume553
DOIs
StatePublished - 2015

Fingerprint

Riboswitch
RNA
Sampling
Nucleotide Motifs
Topology
Data Mining
Trees (mathematics)
Torsional stress
Data mining

Keywords

  • Hierarchical graph sampling
  • RAG
  • RAGTOP
  • Riboswitch
  • RNAJAG
  • Structure prediction

ASJC Scopus subject areas

  • Biochemistry
  • Molecular Biology

Cite this

Computational prediction of riboswitch tertiary structures including pseudoknots by RAGTOP : A hierarchical graph sampling approach. / Kim, Namhee; Zahran, Mai; Schlick, Tamar.

In: Methods in Enzymology, Vol. 553, 2015, p. 115-135.

Research output: Contribution to journalArticle

@article{b7a1e9dd361841e4a65bc98551a40b20,
title = "Computational prediction of riboswitch tertiary structures including pseudoknots by RAGTOP: A hierarchical graph sampling approach",
abstract = "The modular organization of RNA structure has been exploited in various computational and theoretical approaches to identify RNA tertiary (3D) motifs and assemble RNA structures. Riboswitches exemplify this modularity in terms of both structural and functional adaptability of RNA components. Here, we extend our computational approach based on tree graph sampling to the prediction of riboswitch topologies by defining additional edges to mimick pseudoknots. Starting from a secondary (2D) structure, we construct an initial graph deduced from predicted junction topologies by our data-mining algorithm RNAJAG trained on known RNAs; we sample these graphs in 3D space guided by knowledge-based statistical potentials derived from bending and torsion measures of internal loops as well as radii of gyration for known RNAs. We present graph sampling results for 10 representative riboswitches, 6 of them with pseudoknots, and compare our predictions to solved structures based on global and local RMSD measures. Our results indicate that the helical arrangements in riboswitches can be approximated using our combination of modified 3D tree graph representations for pseudoknots, junction prediction, graph moves, and scoring functions. Future challenges in the field of riboswitch prediction and design are also discussed.",
keywords = "Hierarchical graph sampling, RAG, RAGTOP, Riboswitch, RNAJAG, Structure prediction",
author = "Namhee Kim and Mai Zahran and Tamar Schlick",
year = "2015",
doi = "10.1016/bs.mie.2014.10.054",
language = "English (US)",
volume = "553",
pages = "115--135",
journal = "Methods in Enzymology",
issn = "0076-6879",
publisher = "Academic Press Inc.",

}

TY - JOUR

T1 - Computational prediction of riboswitch tertiary structures including pseudoknots by RAGTOP

T2 - A hierarchical graph sampling approach

AU - Kim, Namhee

AU - Zahran, Mai

AU - Schlick, Tamar

PY - 2015

Y1 - 2015

N2 - The modular organization of RNA structure has been exploited in various computational and theoretical approaches to identify RNA tertiary (3D) motifs and assemble RNA structures. Riboswitches exemplify this modularity in terms of both structural and functional adaptability of RNA components. Here, we extend our computational approach based on tree graph sampling to the prediction of riboswitch topologies by defining additional edges to mimick pseudoknots. Starting from a secondary (2D) structure, we construct an initial graph deduced from predicted junction topologies by our data-mining algorithm RNAJAG trained on known RNAs; we sample these graphs in 3D space guided by knowledge-based statistical potentials derived from bending and torsion measures of internal loops as well as radii of gyration for known RNAs. We present graph sampling results for 10 representative riboswitches, 6 of them with pseudoknots, and compare our predictions to solved structures based on global and local RMSD measures. Our results indicate that the helical arrangements in riboswitches can be approximated using our combination of modified 3D tree graph representations for pseudoknots, junction prediction, graph moves, and scoring functions. Future challenges in the field of riboswitch prediction and design are also discussed.

AB - The modular organization of RNA structure has been exploited in various computational and theoretical approaches to identify RNA tertiary (3D) motifs and assemble RNA structures. Riboswitches exemplify this modularity in terms of both structural and functional adaptability of RNA components. Here, we extend our computational approach based on tree graph sampling to the prediction of riboswitch topologies by defining additional edges to mimick pseudoknots. Starting from a secondary (2D) structure, we construct an initial graph deduced from predicted junction topologies by our data-mining algorithm RNAJAG trained on known RNAs; we sample these graphs in 3D space guided by knowledge-based statistical potentials derived from bending and torsion measures of internal loops as well as radii of gyration for known RNAs. We present graph sampling results for 10 representative riboswitches, 6 of them with pseudoknots, and compare our predictions to solved structures based on global and local RMSD measures. Our results indicate that the helical arrangements in riboswitches can be approximated using our combination of modified 3D tree graph representations for pseudoknots, junction prediction, graph moves, and scoring functions. Future challenges in the field of riboswitch prediction and design are also discussed.

KW - Hierarchical graph sampling

KW - RAG

KW - RAGTOP

KW - Riboswitch

KW - RNAJAG

KW - Structure prediction

UR - http://www.scopus.com/inward/record.url?scp=84923532173&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84923532173&partnerID=8YFLogxK

U2 - 10.1016/bs.mie.2014.10.054

DO - 10.1016/bs.mie.2014.10.054

M3 - Article

VL - 553

SP - 115

EP - 135

JO - Methods in Enzymology

JF - Methods in Enzymology

SN - 0076-6879

ER -