-
Notifications
You must be signed in to change notification settings - Fork 1
Installing BioWardrobe
This is the installation documentation for BioWardrobe 1.0.0 and higher.
Hardware requirements will depend on the largest genome you are going to map to. BioWardrobe can be run on higher end consumer class hardware. For example, our virtual machine using 4 cores and 6 Gb of RAM can process ~10M reads of drosophila ChIP-Seq within ~ 2 hrs on an office computer with i7 and 8Gb of RAM. Bigger genomes require higher amounts of RAM. Another BioWardrobe instance is run on a Mac Pro with 64Gb of RAM and attached Thunderbolt HDD array. It is more than enough to process human and mouse ChIP/RNA-seq data. Our main server has 4 E5 processors with 256 Gb of RAM and is attached to a 16Tb 1Gb NAS array.
Wardrobe should run on most Unix, Linux, Mac and Mac Server systems. This time only OpenSUSE 13.2 and MacOS X Yosemite are supported.
- MariaDB 10
- MySQL 5.6
List of required libraries and packages that need to be installed before installing BioWardrobe: Installation Requirements.
I assume that default directory for BioWardrobe is /wardrobe
and BioWardrobe sources are in /wardrobe/src
dir.
Get BioWardrobe’s source code by either downloading a packaged release or checking out the code repository. Create a separate directory for BioWardrobe /wardrobe
and clone github repository into /wardrobe/src
mkdir -p /wardrobe/src git clone https://github.com/portah/biowardrobe.git /wardrobe/src
Run install.sh
from the /wardrobe/src
dir with root
credentials.
cd /wardrobe/src
sudo ./install.sh
Install.sh script will:
Create next directory structure:
/wardrobe
|_ ANL-DATA
|_ RAW-DATA
|_ src
|_ bin
|_ ems
|_ indices
|_ gtf
|_ tmp
|_ upload
Compile executables at /wardrobe/src/src/
:
- source:src/atdp
- source:src/iaintersect
- source:src/bam2bedgraph
- source:src/reads-counting
and link the executables and all python scripts into /wardrobe/bin/
Install script makes all necessary changes to apache. To tune them or troubleshoot read Apache Config