ParGenes master

Basic information

Tested on (Requirements)

  • OS base: CentOS (x86_64) \(\boldsymbol{\ge}\) 6.6 (Rocks 6.2)
  • Dependencies to run ParGenes:
    • GCC 5.0 > (tested on version 5.4.0)
    • Python 2.7 or 3.X (tested on Python 3.6.5)
    • CMAKE > 3.6 (tested on version 3.7.1)
    • mpich2 (tested on version 3.2)

Installation

The following procedure is the easiest way to install ParGenes (master git version) in a cluster.

  1. Clone the git repository of ParGenes in a cluster location (we are going to use the $HOME directory).

    $ git clone --recursive https://github.com/BenoitMorel/ParGenes.git

    Note

    ParGenes at the moment only runs via .py files, therefore is not posible to install it properly in a cluster. Only the user that clones the repo can run ParGenes.

  2. Load the dependencies so ParGenes will be able to run, and check the GCC version.

    $ module load mpich2/3.2_gcc-5.4.0 python/3.6.5_miniconda-4.5.1 cmake/3.7.1
$ gcc --version
gcc (GCC) 5.4.0

  3. Now, for being able to run ParGenes please follow these commands:

    $ cd ParGenes
$ ./install.sh

    Now you have ParGenes installed inside your $HOME directory.

Running Example

In this section, there is an example run that ParGenes already has.

  1. First, we create a conda environment, so we can run ParGenes in a secure environment (it is not necessary but we recommend it).

    $ conda create --name pargenes
$ conda activate pargenes
$ cd examples/data/small/
$ python ../../../pargenes/pargenes-hpc.py -a fasta_files/ -o output_dir -c 32 -d nt -R "--model GTR"

    Note

    Make sure to load every module from the beginning, specially mpich2.

    If everything runs without errors, you have installed ParGenes successfully in your $HOME directory

  2. This is an example for Apolo in SLURM.
    #!/bin/bash

#SBATCH --partition=longjobs
#SBATCH --nodes=2
#SBATCH --ntasks-per-node=1
#SBATCH --cpus-per-task=16
#SBATCH --time=1:00:00
#SBATCH --job-name=ParGenes
#SBATCH --partition=longjobs
#SBATCH -o result_%N_%j.out
#SBATCH -e result_%N_%j.err

module load mpich2/3.2_gcc-5.4.0 python/3.6.5_miniconda-4.5.1 cmake/3.7.1

python ~/scripts/ParGenes/pargenes/pargenes-hpc.py -a ~/Bacillus_subtilis/ParGenes_data/mix_msa -o ~/output_dir_slurm3 -c 32 -d nt -b 10000 -R "--model GTR"

For more information on how to use ParGenes, please visit the official website.

References

ParGenes - ParGenes Official website.
Retrieved Octubre 4, 2019, from https://github.com/BenoitMorel/ParGenes/wiki
Installation - ParGenes Official Website.
Retrieved Octubre 4, 2019, from https://github.com/BenoitMorel/ParGenes#installation

Authors