Bioinformatics Database "Package Manager".

# List avaiable databases
darwin@beagle> pipet list
ncbi/nt

# Download a database
darwin@beagle> pipet pull ncbi/nt
Downloading ncbi/nt.fasta from ftp://ftp.ncbi.nih.gov/blast/db/fasta/nt.gz
Success!

# on most 'nix systems
(sudo) gem install pipet

Databases are defined using Ruby. However, the Domain-Specific Language should make it possible to define databases without having to know how to program in Ruby:

class NCBINT < Database

  name 'ncbi/nt'
  url 'ftp://ftp.ncbi.nih.gov/blast/db/FASTA/nt.gz'
  description 'NCBI non-redundant nucleotide database'
  md5 'f77aff452dd1ba6b0e26b8d6b97b0a62'

end

See Sources/ for examples.

1. I got tired of having to hunt for databases and thought it would be useful to store their locations and md5 checksums all in one place.

2. Most bioinformatics databases do not use any sort of versioning which hinders reproducibility of computation experiments. Pipet uses a combination of MD5 checksums and Git which allows for a type of versioning (the original database may no longer be available on the web but at least we can tell).

3. Homebrew made my life pretty awesome.

Copyright (c) 2013 Austin Richardson. See LICENSE.txt for details.