ModeRNA: a tool for comparative modeling of RNA 3D structure

Abstract

RNA is a large group of functionally important biomacromolecules. In striking analogy to proteins, the function of RNA depends on its structure and dynamics, which in turn is encoded in the linear sequence. However, while there are numerous methods for computational prediction of protein three-dimensional (3D) structure from sequence, with comparative modeling being the most reliable approach, there are very few such methods for RNA. Here, we present ModeRNA, a software tool for comparative modeling of RNA 3D structures. As an input, ModeRNA requires a 3D structure of a template RNA molecule, and a sequence alignment between the target to be modeled and the template. It must be emphasized that a good alignment is required for successful modeling, and for large and complex RNA molecules the development of a good alignment usually requires manual adjustments of the input data based on previous expertise of the respective RNA family. ModeRNA can model post-transcriptional modifications, a functionally important feature analogous to post-translational modifications in proteins. ModeRNA can also model DNA structures or use them as templates. It is equipped with many functions for merging fragments of different nucleic acid structures into a single model and analyzing their geometry. Windows and UNIX implementations of ModeRNA with comprehensive documentation and a tutorial are freely available.

Figure 1.

Distances used in the pre-filtering stage of the fragment search.

Figure 2.

Evaluation of tRNA models generated from templates and alignments—relation between the sequence identity and (a) the GDT_TS score, (b) all atom RMSD of models against experimentally solved structures, (c) the P atom and the C4′ atom RMSD of models against experimentally solved structures (in black) and experimentally solved structures (templates) against experimentally solved structures (targets) (in red) and (d) all atom RMSD of models against experimentally solved structures where the anticodon and the CCA regions are excluded.

Figure 3.

Examples of models built by ModeRNA. Models are shown in red, the native structure is shown in green. (a) Model of E. coli tRNA_Phe (native structure 2J00_W) built on the target 2HGP_D (E. coli tRNA_Phe). The sequences of both molecules are 100% identical, the RMSD value is relatively high—3.61. The residues that contribute the most in the high RMSD are marked with gray clouds and their conformation is shown in separate boxes. (b) Model of E. coli tRNA_Thr (the native structure 1QF6_B—PDB-ID 1QF6, chain B) built on the template 1B23_R E. coli tRNA_Cys. The native structure is interacting with threonyl-tRNA synthetase, while the template structure is in contact with the translation elongation factor EF-Tu-shifting the conformation of the acceptor stem and anticodon loop by several Å, both regions are marked by gray clouds. (c) Model of E. coli tRNA_fMet (native structure 2HGI_C) built on the template 2B64_V (E. coli tRNA_Phe). Model structure has low RMSD—1.38 Å despite medium sequence similarity (47%) between the target and the template molecule. (d) Model of tRNA_Glu (native structure 2DXI_C) built on 2DET. Both structures have a high-sequence similarity (72%). Yet, the RMSD amounts 8.05 Å. The reason is the 6 nt long fragment that is missing in the template on the 3′ end. In the model it has a completely different conformation than the native one. (e) Adenine-binding riboswitch (1Y26) modeled using a guanine-binding riboswitch (1Y27). (f) 30S ribosomal subunits from T. thermophilus (1J5E_A) modeled using 30S from E. coli (2AVY_A). Three regions where the model did not match the native PDB structure well are highlighted: I—two hairpins connected by a junction (residues 970–1022), II—stem loop (residues 65–89), III—stem loop (residues 173–196).

Figure 4.

Evaluation of tRNA models generated from templates and alignments with (a) DI and (b) DP measures.

Figure 5.

The model of Azoarcus group I intron built with ModeRNA (in red) and with RNABuilder (51) (in blue) compared with the experimentally solved structure, PDB code 1U6B (in green).