RNA 3D design: rna_design

Metadata

Author: Rhiju Das

This document updates documentation written in 2008 by Rhiju Das (rhiju [at] stanford.edu) into the latest documentation format. Last update: April 2011.

Code and Demo

The central code for the rna_design application is in apps/public/rna/rna_design.cc with core routines run through the amazing Rosetta packer.

For a 'minimal' demo example of RNA design:

demos/public/RNA_Design [in the release version]

References

Das, R., Karanicolas, J., and Baker, D. (2010), "Atomic accuracy in predicting and designing noncanonical RNA structure". Nature Methods 7:291-294. [for high resolution refinement] Paper. Link.

(Reprint available at http://daslab.stanford.edu/pubs.html ).

Purpose

This code is intended to carry out fixed backbone design of RNA sequences given an input backbone.

Algorithm

This application carries out combinatorial optimization of nucleobase type and conformation along with 2'-OH torsions, in the context of a pre-specified RNA backbone. It is very similar to the Rosetta fixed-backbone protein design algorithm, and has been used to test the new Rosetta RNA potential. Unfortunately, it is not presently very optimized for speed; the precalculation of rotamer energies takes a while. Runs on RNA backbones longer than ~ten nucleotides take many minutes or hours; algorithm improvements implemented in future releases will greatly speed this up.

Limitations

Input Files

Required file

Just the PDB file with desired backbone.

How to run with this file.

rna_design.<exe> -s chunk001_uucg_RNA.pdb   -nstruct 3  -ex1:level 4 -dump -score:weights rna_hires.wts -database <path to database>

This demo redesigns a 'UUCG' tetraloop on a single-base pair RNA 'helix', as a small 6-nucleotide test case. As illustration, only 3 designs are output. It takes about 15 seconds to run. The typical sequence output is cuuggg (native is cuucgg).

Options

-database                                        Path to Rosetta databases. [PathVector]
-in:file:s                                       Name(s) of single PDB file(s) to process. [FileVector]
-nstruct                                         Number of times to process each input PDB. [Integer]
-ex1:level <n>                                   Use extra chi1 sub-rotamers for all residues that pass
                                                 the extrachi_cutoff.
                                                 [Boolean]
                                                 The integers that follow the ex flags specify the pattern
                                                 for chi dihedral angle sampling.
                                                 There are currently 8 options; they all include the original
                                                 chi dihedral angle No. 4 means: EX_TWO_HALF_STEP_STDDEVS
                                                 [-1,-1/2,0,1/2,1 standard deviations].
-dump                                            Generate pdb output,default:false. [Boolean]
-score:weights rna_hires.wts                     Name of weights file, default is standard. [String]
-sample_chi                                      Sample chi (glycosidic torsion angle).
-disable_o2star_rotamers                         Turn off sampling of 2'-OH proton position.

Tips

What do the scores mean?

The most common question we get is on what the terms in the 'SCORE lines' of silent files mean. Here's a brief rundown, with more explanation in the papers cited above.

***Energy interpreter for fullatom silent output:
score                                            Final total score
fa_atr                                           lennard-jones attractive
fa_rep                                           lennard-jones repulsive
fa_intra_rep                                     Lennard-jones repulsive between atoms in the same residue
lk_nonpolar                                      lazaridis-karplus non-polar solvation energy
hack_elec_rna_phos_phos                          Simple electrostatic repulsion term between phosphates
hbond_sr_bb_sc                                   backbone-sidechain hbonds close in primary sequence
hbond_lr_bb_sc                                   backbone-sidechain hbonds distant in primary sequence
hbond_sc                                         sidechain-sidechain and sidechain-backbone hydrogen bond energy
ch_bond                                          Carbon hydrogen bonds
geom_sol                                         Geometric Solvation energy for polar atoms
rna_torsion                                      RNA torsional potential.
atom_pair_constraint                             Atom pair distance constraints score?
angle_constraint                                 (not in use)
rms                                              rmsd

Expected Outputs

If you use the sample flag files, there are also other output files generated.

start.pdb:                                       Idealized structure
S_000*.pdb:                                      Output of the rna denovo design.
chunk001_uucg_RNA.sequence_recovery.txt:         This is a simple report for design identity of each RNA residue and structure.
chunk001_uucg_RNA.pack.txt:                      Total score and sequence for each output model
chunk001_uucg_RNA.pack.out:                      Scores (with breakdown by score component) for each re-designed sequence