C-RCPred: Constrained RNA Complex Prediction

C-RCPred is an open-source bioinformatics program for the prediction of RNA complexes.
The multi-objective algorithm is implemented in C++ and runs on UNIX-like operating systems.

Besides this stand-alone version, a web server is freely available on the EvryRNA platform.

Installation

If all the dependencies are already installed on your system, using the make command within the C-RCPred directory is sufficient to build the executable file.
Alternatively, C-RCPred can easily be installed by using Docker as follows:

git clone https://forge.ibisc.univ-evry.fr/EvryRNA/C-RCPred.git
cd C-RCPred
docker build -t c-rcpred ./

The above docker build command has been tested for Docker version 20.10.7.

Dependencies

The installation of C-RCPred depends on two bioinformatics libraries:

• the ViennaRNA package (version 2.5.1)¹
• the NUPACK software suite (version 3.2.2)²

The CPLEX Optimizer (version 22.10) is also required.

In the Dockerfile, the other dependencies that must be installed can be found.

Testing

Two sets of files are provided to test C-RCPred. Each set contains 4 input files (.fasta, .txt30, .txt90, and .txt).
Thus, an example of command is:

./C-RCPred -f test/NDB_00016.fasta \
           -s test/NDB_00016_str.txt30 \
           -i test/NDB_00016_inter.txt90 \
           -p test/NDB_00016_probing.txt \
           -l 20 -u 80 -t 100 -e 1 \
           -g NDB_00016_graph.txt \
           -q NDB_00016_cliques.txt -m 20

This command will output predicted secondary structures, represented with the dot-bracket notation.
Other results will be written in the NDB_00016_graph.txt and NDB_00016_cliques.txt files (see options below).

A similar test can be performed with the files provided for PDB_00805.

Options

The following list of options can be accessed by running ./C-RCPred -h:

 -f       Input fasta file
 -s       Input secondary structure files,
 -i       Input interaction secondary structure files,
 -c       Constraint files where constraints are in the same order than the secondary structure and interaction files
 -p       Probing data file for each RNA, in the same order than the fasta files (if no data for an RNA, an empty file is needed)
 -o       Format of output "d" (dot-parenthesis, default), "j" (JSON) or "b" (base pair list format)
 -e       Energy model (0: No energy model; use energies computed by upstream tools, 1: ViennaRNA package model; default, 2: NUPACK model)
 -t       Threshold for compatibility RNA (0->100, default 100)
 -l       Lower probing threshold (0->100, if not specified, probing values between 0% and the upper threshold once normalized will not be taken into account)
 -u       Upper probing threshold  (0->100, if not specified, probing values between the lower threshold and 100% once normalized will not be taken into account)
 -g       Output graph file
 -k       Output cliques predicted
 -q       Output cliques predicted without duplicates
 -m       Maximum number of structures to output

Contact

✉️ fariza.tahi@univ-evry.fr

References