Join GitHub today
GitHub is home to over 28 million developers working together to host and review code, manage projects, and build software together.Sign up
OPAL is optionally available as a OCTAVE package (open-source package). Please click here to download. In this package OCTAVE is used instead of MATLAB.
#Installation To install OPAL on UNIX system
- Unpack: tar -zxvf OPAL.tar.gz
- cd OPAL
#Installing the support packages:
- OPAL requires installation of SPIDER2 package, BLAST and its databases, MoRFchibi package, MATLAB/OCTAVE, GAWK and other basic compilers:
1. Installing SPIDER2 package
SPIDER2 package can be downloaded from "http://sparks-lab.org/pmwiki/download/index.php?Download=yueyang/SPIDER2-hse_local.tgz"
unpack: tar xzvf SPIDER2-hse_local.tgz
Install Blast: SPIDER2 requires BLAST+ software package to be installed with NR database. BLAST+ can be downloaded from NCBI: https://blast.ncbi.nlm.nih.gov/Blast.cgi?PAGE_TYPE=BlastDocs&DOC_TYPE=Download. Alternatively, BLAST+ can be installed on Ubuntu system using command line: sudo apt-get install ncbi-blast+
NR dataset: you can download them from "ftp://ftp.ncbi.nlm.nih.gov/blast/db/"
Note: Please in file "/SPIDER2_local/misc/run_local.sh" change the blastpgp & NR database path to correct addresses.
e.g. blastpgp=/usr/bin/psiblast and NR=/home/pub/blast/NR/nr
2. Installing MoRFchibi package
- MoRFchibi can be downloaded from "http://www.chibi.ubc.ca/wp-content/uploads/2016/03/MCS1.03.tar.gz"
- tar -zxvf MCS1.03.tar.gz
- cd MCS1.03
- Compile: g++ -O3 source/mc.c source/*.cpp -o mc
3. Installing MATLAB and LibSVM tool box
If using MATLAB package, download and install MATLAB from Mathworks OR if using OCTAVE package, install OCTAVE by command line: sudo apt-get install octave.
OPAL uses LibSVM tool (Chang and lin 2011) for classification task, LibSVM package is available at the download page: http://www.csie.ntu.edu.tw/~cjlin.
Note: LibSVM MATLAB/OCATVE binaries are required to be built using make file and its path to be included in “properties.txt’ file in OPAL directory.
4. Installing other compilers using command line:
sudo apt-get install gawk
sudo apt-get install python
sudo pip install scipy
sudo apt-get install python-scipy
sudo pip install -I numpy==1.11.1
To use OPAL, first all the required compilers and packages mention above must be installed. Secondly, “properties.txt’ file in OPAL directory must be edited as follows:
- Path to spider2_local misc dir must be adjusted
- Path to LibSVM MATLAB binaries must be adjusted
- Path to PROMIS dir must be adjusted
- Path to MoRFchibi MCS1.03 dir must be adjusted
To run OPAL predictor:
- The query sequence in fasta format should be kept in OPAL directory as input.fasta
- if pssm file is available for query sequence, then should be kept in OPAL directory as input.pssm
For demo: input sequence and pssm files are available in OPAL directory for protein sequence of length 207
run_OPAL.m file can be used to predict MoRF scores
- e.g. run_OPAL('sequence'); #input: query sequence
- or e.g. run_OPAL('sequence','pssm'); #input: query sequence and pssm file if available (if pssm file available, it will not be computed)
The input query file will be processed and results will be saved as scores_OPAL.txt file
Note: To call run_OPAL function, Demo.m file is given. If you are using OCTAVE graphical user interface you can open Demo.m and simulate. This will run run_OPAL function with default input as sequence only. If you are calling from command line, you can type 'sudo octave --eval Demo'.
Alternatively, PROMIS predictor can be also used similar way:
- e.g. run_PROMIS('sequence'); #input: query sequence
- or e.g. run_PROMIS('sequence','pssm'); #input: query sequence and pssm file if available
The input query file will be processed and results will be saved as scores_PROMIS.txt file
Demo.m file is given as a demo to run predictors OPAL and PROMIS using either MATLAB or OCTAVE.