Reference

• J. Sperschneider, A. Datta and M.J. Wise. Predicting pseudoknotted structures across two RNA sequences. Bioinformatics. 2012.

Download

• Code [DotKnot_PW_1.1.tar.gz]

Testset

• Sequences and structures used for evaluation in the paper [TestSet_RFAM.tar.gz]

Results

• Results from the evaluation in the paper [TestSet_RFAM_Results.tar.gz]

DotKnot-PW: Predicting pseudoknotted structures across two RNA sequences

DotKnot-PW 1.1 has been written in Python and uses RNAfold and RNAeval from the Vienna RNA Package for producing the probability dot plot and for free energy evaluation of secondary structure elements. To get DotKnot-PW to work on your local machine, you need to get Python, SciPy and the Vienna RNA Package running.

DotKnot-PW has been tested on Linux, Windows using Cygwin and Mac OS.

Installation

(1) Unpack DotKnot_PW_1.1.tar.gz in your desired location:

tar -zxvf DotKnot_PW_1.1.tar.gz

(2) You need to install Python and SciPy to run DotKnot-PW, see https://www.python.org and https:///www.scipy.org.

(3) You need to install the Vienna RNA package from source. The Vienna RNA package is already provided in the DotKnot-PW folder for one reason. The default cutoff in RNAfold for the base pair probabilities which are displayed in the dotplot is 1E-5, however for pseudoknot prediction we need a much lower cutoff of 1E-11. In the Vienna RNA package which comes with DotKnot-PW, this cutoff has already been changed by hand in the C source code.

You need to switch to the directory and unpack, configure and make:

cd DotKnot_PW_1.1/scr
tar -zxvf ViennaRNA-1.8.5.tar.gz
cd ViennaRNA-1.8.5
./configure
make
and (as root)
make install

Usage

Switch to the DotKnot-PW working directory:

cd DotKnot_PW_1.1/scr

DotKnot-PW needs to be called with an input FASTA file:

python dotknot.py [input_file]

For example:

python dotknot.py ../TestSet_Rfam/RF00094_HDV/RF00094_10seqs.fasta