Generate quick secondary structures of nucleic acids via jupyter notebook or command line.
A communal and continuing effort by the Das Lab, Eterna Players, M. Wu, H. Wayment-Steele.
The default behavior is to output the secondary structure visualization to both svg and png. The png requires Inkscape, whose directory must be set by environmental variable INKSCAPEDIR.
Clone this repository to your hard drive and install as follows:
git clone [email protected]:DasLab/draw_rna.git
cd draw_rna/
python setup.py install
Warning if you see errors in using the repository, it maybe due to your draw_rna directory sitting in a folder like src/ which already is in your Python Path. in that case, rename draw_rna to draw_rna_directory and/or move it to another spot in your system.
See example_files/demo.ipynb for more. In brief:
An input file is used to specify the sequence, secondary structure, and coloring of the desired drawing. The format is as follows
filename # image will be written to filename.svg
GTGANNNNNTCAC # nucleic acid sequence
((((.....)))) # nucleic acid secondary structure in dot-bracket notation
rrrrbbbbbgggg # optional: coloring of each base (e.g. r for red, g for green, b for blue)
Multiple sequences can be specified in one file as in example_input.txt.
The draw script can be run with
python draw_rna/draw_all.py example_files/command_line/example_input.txt
You'll get files example_contours.svg, example_sequence_colors.svg, example_specific_colors.svg. Example output available in example_files/command_line/.
At the moment, current practice is to include blanks between strands such as:
filename
GTGANN NNTCAC
((((.....))))
rrrrbbwbbgggg
And then remove nucleotides corresponding to blanks afterwards. (This would be a quick fix.)
'render_rna_flip.py' is what is currently called in the main routine and where major edits have been made. render_rna.py renders 3' -> 5'. TODO: handle 3' -> 5' drawing better.
python draw_all.py covid19_5primeUTR.txt --large_mode --color_values perc_conserved_5UTR.txt --png
If you use --color_values with an input file with multiple structures, the same colors will be applied to all the structures.
Requires RDATkit to be in working python path. Example command:
python draw_from_rdat.py ETERNA_R69_0000.rdat 297 --png
More coming eventually. For now, if you generate a bunch of stills, say from a cotranscriptional simulation, can string them together with ImageMagick:
convert -delay 25 my_stills_*.png my_great_trajectory.gif