Installation¶
Dependency installation¶
Mandatory:
- samtools-v1.3.1
- BWA-v0.7.12
- MACS2-v2.1.1.20160309
- Picard
- java-v1.8 (required by Picard)
Optional:
- igraph-v0.7.1 C library. (This is required if you want to compile the program from the source)
- STAR-v2.5.2b (RNA-seq analysis)
- RSEM-1.2.31 (RNA-seq analysis)
Warning
Software with older versions might not work.
Software installation¶
Note
Before trying to install the program from source, you should check whether a binary release is available for your platform. If the answer is “Yes”, simply download the binary file, put it in your system path and skip the following sections. Otherwise, continue to read.
Install stack¶
Download the latest release of stack for your platform.
Note
Older Linux distribution may still use libgmp4
, i.e, Centos 6.x.
If that is the case, you need to download the version compiled with libgmp4
.
For example, something like “Linux 64-bit, libgmp4 for CentOS 6.x”.
Next, unpack the tarball and move stack
executable to a directory
that is in your PATH
environment variable, e.g., /usr/bin
.
tar zxf stack-x.x.x-linux-x64.tar.gz
mv stack-x.x.x-linux-x64/stack /usr/bin
Download Taiji source and install GHC¶
Download the latest source code and unpack it.
tar zxf Taiji-X.X.X.tar.gz
Go into the source code directory and install GHC.
cd Taiji-X.X.X
stack setup
Once you have a working copy of GHC, you can proceed to install the dependencies (This will take a long time for the first time). Under the source code directory, type:
stack build --only-dependencies
and then install Taiji:
stack install
Or
stack install --flag SciFlow:sge
If you want SGE integration. (This requires that the DRMAA C library is installed.)
After the installation, the taiji
executable should be in /home/yourname/.local/bin
.
Please add this directory to the PATH
environment variable (Do not know how to do this? See Linux Primer).