ILRI Research Computing

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
====== SpeedSeq ======
A flexible framework for rapid genome analysis and interpretation.

===== Information =====

  * Version: 0.1.2
  * Added: June, 2017
  * Link: https://github.com/hall-lab/speedseq

===== Usage =====

See versions of speedseq which are available:
<code>$ module avail speedseq</code>

Load and run speedseq:
<code>$ module load speedseq/0.1.2
$ speedseq</code>

===== Installation ======
Notes from the sysadmin during installation:

<code>$ sudo mkdir -p /export/apps/speedseq/0.1.2
$ sudo chown aorth /export/apps/speedseq/0.1.2
$ git clone -b v0.1.2 --recursive https://github.com/hall-lab/speedseq.git /export/apps/speedseq/0.1.2
$ cd /tmp
$ wget https://www.python.org/ftp/python/2.7.13/Python-2.7.13.tar.xz
$ tar xf Python-2.7.13.tar.xz
$ cd Python-2.7.13
$ ./configure --prefix=/export/apps/speedseq/0.1.2 --enable-shared --with-threads --with-computed-gotos --enable-optimizations --with-lto --enable-ipv6
$ make -j4
$ make install
$ make maninstall
$ export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/export/apps/speedseq/0.1.2/lib
$ export PATH=/export/apps/speedseq/0.1.2/bin:$PATH
$ export PYTHONHOME=/export/apps/speedseq/0.1.2
$ curl https://bootstrap.pypa.io/get-pip.py | python
$ pip install numpy
$ pip install pysam
$ pip install scipy
$ cd /export/apps/speedseq/0.1.2
# Remove -lCint from ROOTLIBS in src/cnvnator/Makefile
# Add -std=c++11 to CXX in src/cnvnator/Makefile
# See: https://github.com/abyzovlab/CNVnator/issues/9
# See: https://github.com/hall-lab/speedseq/issues/77
$ module load root/6.08.06
$ make
$ sudo chown -R root:root /export/apps/speedseq</code>

Note: compiled on CentOS 7 because it needs quite a bit of modern dependencies, so make sure to run it from a compute node that is running CentOS 7!

Also, because speedseq requires a Python with several modules installed, I decided to install Python into the speedseq root so that speedseq could have a dedicated Python. This is the reason the install above is slightly out of order (cloning the speedseq dir first, then installing the Python requirement into it).