This program is a slightly modified version of the GT-Scan. In this version, multiple sequences in a single fa file can be searched simultaneously, depending on the number of CPU cores. The original version can be found Here.
#1. First, download and add bowtie to your $PATH variable in $HOME/.bashrc.
export PATH=$PATH:/your/bowtie/path
#2. On the bowtie website, download the pre-built indexes for H. sapiens, UCSC hg19 from the right sidebar.
#3. Clone the repository from this page by using the following command.
git clone https://github.com/thepikabone/MGT-Scan
#4. Enter MGT-Scan directory and and unzip the pre-built indexes from step 2 into the genome directory. Remember this is your genome path.
cd MGT-Scan
cd genome
unzip /your/downloaded/file
#5. Go back into MGT-Scan directory and enter gt-scan directory.
cd ..
cd gt-scan
#6. Edit config.ini and specify the path (ref_genome_dir) to the your genome path from step 4.
ref_genome_dir=/your/genome/path
#7. Set permission of gt-scan.py in the gt-scan directory to 700.
chmod 700 gt-scan.py
#8. Export gt-scan directory to environment variable in $HOME/.bashrc.
export $PATH=/your/gt-scan/path:$PATH
#9. Make a directory called gtscan.db in /tmp. This is where all the temporary files are stored. Delete all files in this directory periodically.
cd /tmp
mkdir gtscan.db
#10. Go back into the MGT-Scan directory, in the console, type R --args , followed by the following arguments:
a. Directory: Directory that contains the .fa file.
b. FileName: Name of the .fa file.
c. n: Numbers of pairs of target candidates per sequence.
d. dis: The minimum distance between individual target candidate in a pair.
e. Cores: Number of CPU cores this program can use.
Separate each argument with a space, then hit enter when you’re finished.
For example:
Example: R --args /home/zw355/ModifiedGT-Scan test.fa 3 80 4
#11. Run GTScan.R and wait for results. The results should be a tab file with the same name as the .fa file.
source("GTScan.R")
#12. Clear temporary .db files in /tmp/gtscan.db after the program is finished running.
The result, shown as a .tab file, may look like this:
chr1 714299 714318 - CGCCCGGCGCCGAAGACCGG chr1 714026 714045 + CCAACGGCCCACCTCTATGG
chr1 714299 714318 - CGCCCGGCGCCGAAGACCGG chr1 714026 714045 + CCAACGGCCCACCTCTATGG
Each line represents a pair of candidate sequences with their chromosome numbers, locations, and directions. The distance between each two should be greater than the given minimum distance. These target candidates are likely to have few to no mismatches.
A reminder that a sequence in the fafile may produce a maximum of n pairs, but it’s possible to have no output.