Contents

Prepare RNA motif

Prepare cleaned RNA motif from PDB file

We are going to use the PDB entry 3P59, an RNA square, as an example. The conformation of the bulge motif formed by strands E and F is in an alternative conformation. We will first try to build an RNA triangle use the EF motif. In Chimera, trim the RNA structure so that only residues 50.E to 60.E, and 108.F to 115.F are left. Save the PDB of the new model. Now, the bulge motif is flanked by a 2-bp helix and a 3-bp helix on either side. This is problematic, because nanotiler1. needs a helix of a minimum of 3-bp to recognize the helix.

Align helix onto the motif (only for motif flanked by less than 3-bp helix)

To fix this, in Chimera, build an ideal 3-bp RNA A-form helix as a new model. Then align the newly built helix onto the 2-bp helix of the cleaned bulge model by the Chimera command:

match #1:2-3.A,1-2.B@c3',c4',c5' #0:112-113.F,50-51.E@c3',c4',c5'

Combine the two models, and save the PDB as bulge_EF_aligned_helix.pdb.

Construct triangle with nanotiler

Locate the helix ends in nanotiler

We need to find the identifiers of the ends of helices for writing the subsequent nanotiler scripts. With terminal open in the working directories:

nanoscript
import bulge_EF_aligned_helix.pdb findstems=true name=bulge1
exportpdb bulge_EF_aligned_helix-TMP.pdb junction=false
tree hxend
exit

We will get a new PDB file bulge_EF_aligned_helix-TMP.pdb and get output like:

root.bulge1.stems.stemA_B_1.B_3_A_9 1.1.5.1.7 1 BranchDescriptor3D  41.804 -22.104  10.349
root.bulge1.stems.stemA_B_1.A_11_B_1 1.1.5.1.8 1 BranchDescriptor3D  34.526 -15.262  12.313
root.bulge1.stems.stemC_D_1.D_3_C_1 1.1.5.2.7 1 BranchDescriptor3D  42.802 -30.639  17.498
root.bulge1.stems.stemC_D_1.C_3_D_1 1.1.5.2.8 1 BranchDescriptor3D  35.113 -35.754  13.214

Use Chimera to open bulge_EF_aligned_helix-TMP.pdb (note that the numbering of chains and residues has been changed by nanotiler) for the identifiers of the helix ends, which are root.bulge1.stems.stemA_B_1.A_11_B_1 and root.bulge1.stems.stemC_D_1.C_3_D_1. We will use stems.stemA_B_1.A_11_B_1 and stems.stemC_D_1.C_3_D_1 in the following scripts.

Search the optimal lengths of connecting helices with nanotiler script

This searching script will search connecting helices lengths ranging from 12 to 17 bps. Note that we designated stems.stemA_B_1.A_11_B_1 and stems.stemC_D_1.C_3_D_1 as JD1 and JD2 in the script. Run the script with:

# no console output
nanoscript bulgeEF-newHelixScan-NanoTiler.script > bulgeEF-newHelixScan-NanoTiler.out 2>&1

Or:

# with console output
nanoscript bulgeEF-newHelixScan-NanoTiler.script 2>&1 | tee bulgeEF-newHelixScan-NanoTiler.out

The searching results are written in the bulgeEF-newHelixScan-NanoTiler.out output file. To quickly locate the energy (start_score of last helix restraint), use the commands:

grep -e "####" -e "HBP1: " -e "generated_helices=h_root_bulge3_stems_stemC_D_1_C_3_D_1_root_bulge1_stems_stemA_B_1_A_11_B_1_root" bulgeEF-newHelixScan-NanoTiler.out > bulgeEF-newHelixScan-NanoTiler-energy-summary.out

We can see that the optimal length of the connecting helix is 14 (start_score=3430)or 15 (start_score=2193).

Rigid-body energy optimization with nanotiler script

Taking the helix length of 14, we next optimize the placement of the motifs and helices use the optimization script. Note that the final optimization steps (optcycles in the script) should not be too small. Run the commands:

nanoscript bulgeEF_newOpt14-NanoTiler.script 2>&1 | tee bulgeEF_newOpt14-NanoTiler.out

We get the output PDB file bulgeEF_newOpt14-5000000.pdb.

Post-processing of the output PDB file

Load the output file and remove the strands of the aligned helices:

nanoscript
import bulgeEF_newOpt14-5000000.pdb findstems=true name=triangle
tree strand

convertPDB bulgeEF_newOpt14-5000000-post-3.pdb


Coarse-grain simulation using simRNA

Sequence file and secondary structure file

Prepare the sequence file and secondary structure file according to simRNA. Add 5’-phosphate groups if they are lost in the structure.

Motif and sticky-end stacking restraints

To generate motif restraints, first trim the helices flanking the bulge to 1-bp, save the PDB as bulgeEF_1bp_for_restraints.pdb, which can be accessed here. Use the following commands:

cat > motif_restraints
python res_renumber_pdb_atom.py bulgeEF_1bp_for_restraints.pdb 58E 19A 110F 13B
python res_renumber_pdb_atom.py bulgeEF_1bp_for_restraints.pdb 58E 19C 110F 13D
python res_renumber_pdb_atom.py bulgeEF_1bp_for_restraints.pdb 58E 19E 110F 13F
<control+d>
source motif_restraints
ls | grep restraints_bulgeEF_1bp | xargs cat > restraints_motif.dat

For sticky-end stacking restraints,

python /Users/wyssuser/programs/SimRNA_64bitIntel_MacOSX_staticLibs/simrna_scripts-master/get_sticky_from_seq_ss.py bulgeEF_triangle.seq bulgeEF_triangle.ss
ls | grep restraintsSE | xargs cat >> restraints_SEall.dat

Change the weight to 0.2

cat restraints_motif.dat restraints_SEall0.2.dat > Allrestraints_SE0.2.dat

Run simulation with simRNA

ln -s $SIMRNADATA data
SimRNA -c config.dat -p bulgeEF_newOpt14-5000000-post-3.pdb -S bulgeEF_triangle.ss -r Allrestraints_SE0.2.dat -o bulgeEF_triangle_try1 2>&1 | tee bulgeEF_triangle_try1.out


  1. Shapiro. ↩