Skip to content

install.rst

Latest commit

 

History

History
181 lines (107 loc) · 5.54 KB

install.rst

File metadata and controls

181 lines (107 loc) · 5.54 KB

Installation

System Packages

In order to installed developer tools you will need root privileges; that is, somebody who can use su or sudo

CentOS

#> yum install openmpi openmpi-devel git python-devel zlib-devel ncurses-devel freetype-devel libpng-devel wget java-1.6.0 dejavu*
#> yum groupinstall Development Tools

Ubuntu

#> apt-get install openmpi-bin libopenmpi-dev python-dev git zlib1g-dev build-essential libncurses5 libncurses5-dev libpng12-dev libfreetype6-dev

Installation

  1. Clone the repository

    git clone $(eval echo https://$(read -p "Gitub username: " gu; echo $gu)@github.com/VDBWRAIR/pathdiscov.git)
    cd pathdiscov

  2. Setup a :ref:`virtualenv <activate>` to install into and build documentation

    1. Install virtualenv python environment

      wget https://raw.githubusercontent.com/necrolyte2/bootstrap_vi/master/bootstrap_vi.py -O- | python - pathdiscov --prompt="(pathdiscov)"

    2. Activate the virtualenv to install everything into

      source pathdiscov/bin/activate
pip install paver

    3. If you want to view/install the built html documentation

      paver doc_html
firefox docs/build/html/install.html#id1

    4. If you want to view/install the man page documentation

      paver doc_man
mkdir -p pathdiscov/man/man1
cp docs/build/man/* pathdiscov/man/man1
man pathdiscov

  3. Setup pathdiscov/files/config.yaml.base

    mkdir -p ~/tmp  # or change the location  of  `tmp` dir in `pathdiscov/files/config.yaml.base` required for `diamond`
cp pathdiscov/files/config.yaml{.base,}

  4. Edit config.yaml to suite your setup

    vim pathdiscov/files/config.yaml

    Example edits:

    SEQUENCE_PLATFORM: illumina #choices are: illumina,454

  5. Install the pipeline into the virtualenv

    python setup.py install

  6. Databases setup

    You must refer to built documentation to set up these databases. These databases must be built before you can verify below.

    See :doc:`databases` or databases.rst if you have not built the docs

  7. Quick verify of a few things

    • See if required executables are available

      # These should now all be in your path so should work
apps=( bwa samtools bowtie2 blastx blastn Ray Ray2 cutadapt getorf run_standard.pl fastqc prinseq-lite.pl diamond snap pathdiscov_cli)
for p in ${apps[@]}; do $p --help 2>&1 | grep -qiE '\[main\]|usage|useage|qualifiers|DESCRIPTION|Syntax' && echo "$p ok" || echo "$p broken?"; done

    • See if your databases are available as specified in config

      verifydatabases pathdiscov/files/config.yaml

  8. Optional: Run a Paired-end dataset

    Anytime you run the pipeline you need to activate the pipeline first. If the pipeline is activated you will see `(pathdiscov)` in front of your prompt.

    If it is not activated:

    source ~/pathdiscov/pathdiscov/bin/activate
    pathdiscov_cli --R1 testData/F.fastq --R2 testData/R.fastq --outdir testoutDir

    You can check this project against when it was run during development by heading over to :ref:`Inspect Stage Counts<count-files-interpretation>`

Offline Installation

There may be some instances where you need to install onto an offline workstation. You can achieve this by the following method

  1. Clone the pathdiscov project from github

  2. Download all of the required software prior to installation and place in pathdiscov/download

  3. Download all of the required python packages

    mkdir -p pathdiscov/download/python_packages; pip install --no-use-wheel -d pathdiscov/download/python_packages -r requirements-dev.txt
pip install --no-use-wheel -d pathdiscov/download/python_packages virtualenv paver
  4. Once downloaded make sure all of the files are extracted if needed and the following directories/files exist
    • pathdiscov/download/htslib
    • pathdiscov/download/samtools
    • pathdiscov/download/bwa
    • pathdiscov/download/fastqc_v0.11.2.zip

  5. Now you can copy the git cloned pathdiscov directory to your offline workstation to kick off the install

    cd pathdiscov

  6. Install virtualenv and python packages into that virtualenv

    tar xzf pathdiscov/download/python_packages/virtualenv*
python virtualenv*/virtualenv.py pathdiscov
. pathdiscov/bin/activate
pip install --no-index --find-links=pathdiscov/download/python_packages six argparse numpy paver
pip install --no-index --find-links=pathdiscov/download/python_packages -r requirements-dev.txt

  7. Now you can start the normal installation process from the Databases setup step