Johannes Köster

2016

• provide binary packages of any Bioinformatics software
• fast and easy deployment
• no admin rights
• allow simultaneous usage of multiple versions of the same package

Goals

Already over 1600 packages

(C, C++, Python, R, Perl, ...)

Rapidly growing

Over 100 contributors

Vital community

Integrated with 3 popular workflow management systems

Integration with WMS

wget https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh
bash Miniconda3-latest-Linux-x86_64.sh

conda config --add channels r
conda config --add channels conda-forge
conda config --add channels bioconda

Setup

Bioconda is based on the language agnostic Conda package manager

conda install bwa=0.7.15

conda update bwa

conda remove bwa

Isolated environments:

channels:
  - r
  - conda-forge
  - bioconda
dependencies:
  - picard ==2.3.0
  - samtools ==1.3.0
  - bioconductor-deseq2 ==1.10.1

Usage

Recipe with metadata:

package:
  name: bwa
  version: "0.7.15"

source:
  fn: v0.7.15.tar.gz
  url: https://github.com/lh3/bwa/archive/v0.7.15.tar.gz

requirements:
  build:
    - gcc
    - zlib
  run:
    - zlib

test:
  commands:
    - bwa 2>&1 | grep 'index sequences in the'

about:
  home: http://bio-bwa.sourceforge.net
  license: MIT
  summary: The BWA read mapper.

Packaging made easy

Bash script for compilation:

#!/bin/bash

export C_INCLUDE_PATH=${PREFIX}/include
make

mkdir -p $PREFIX/bin
cp bwa $PREFIX/bin

Packaging made easy

Conda can be used to handle data as well.

Watch out for the upcoming

Go Get Data project:

conda install --channel ggd hg38-dbsnp

Beyond software

Visit

https://bioconda.github.io