BioInfo

BioInfo is a reimagining of bioinformatics libraries for Perl6 based on all the new goodies available such as parallel processing through Promises and Channels and well defined and extensible parsing with the use of Grammars. Not to be confused with the bioperl6 project lead by Chris Fields which aims to be a reimplementation of the BioPerl API for Perl6. Checkout Chris' project for any serious business.

The Good

• Being built to work in parallel use cases so you don't have to worry too much about this yourself.
• Using sped up parse as you go forms of Perl6 Grammars so that a whole file isn't read into RAM and you can just play with getting higher level objects quickly.
• I'll attempt to work on reasonable requests for anything, pull requests are welcome just maybe contact me first if you have to make some big architectural decisions to get what you want.
• A fresh project for anyone who has an opinon to get involved pushing that opinion on others.

The Bad

• I'm mostly ignoring BioPerl.
• Like the bioperl6 project I am trying things out so the API is in flux. If you want something production ready, you're going to have to help to get there.

The Ugly

• The author of these modules (Matt Oates) mostly concerns himself with amino acid sequences so other considerations are secondary. Sorry about that.
• bioperl6 does exist. So this is a bit of a split effort, really bioperl6 should get any attention.

An Example

Dealing with sequences from a file:

use BioInfo::Parser::FASTA;
use BioInfo::IO::FileParser;

#Spawn an IO thread that parses the file and creates BioInfo::Seq objects on .get
my $seq_file = BioInfo::IO::FileParser.new(file => 'myseqs.fa', parser => BioInfo::Parser::FASTA);

#Print all the IDs from the sequence file
while (my $seq = $seq_file.get()) {
    say $seq.id;
}

Using the BioInfo slang to seamlessly manipulate sequences in your code. Anything within backticks or grave accents `` will be treated as FASTA sequence data and create an array of sequence objects:

use BioInfo;

my @seqs = `
>ENST00000505673 cds:NOVEL_protein_coding
NGGTGTGGAGCCATAGGCTGTGAAATGTTGAAAAATTTTGCTTTACTTGGTGTTGGCACA
AGCAAAGAGAAAGGAATGATTACAGTTACAGATCCTGACTTGATAGAGAAATCCAACTTA
AATAGACAGTTCCTATTTCGTCCTCATCACATACAGAAACCTAAAAGCTACACTGCTGCT
GATGCTACTCTGAAAATAAATTCTCAAATAAAGATAGATGCACACCTGAACAAAGTATGT
CCAACCACTGAGACCATTTACAATGATGAGTTCTATACTAAACAAGATGTAATTATTACA
GCATTAGATAATGTGGAAGCCAGGAGATACGTAGACAGTCGTTGCTTAGCAAATCTAAGG
CCTCTTTTAGATTCTGGAACAATGGGCACTAAGGGACACACTGAAGTTATTGTACCGCAT
TTGACTGAGTCTTACAATAGTCATCGGGATCCCCCAGAAGAGGAAATACCATTTTGTACT
CTAAAATCCTTTCCAGCTGCTATTGAACATACCATACAGTGGGCAAGAGATAAGTTTGAA
AGTTCCTTTTCCCACAAACCTTCATTGTTTAACAAATTTTGGCAAACCTATTCATCTGCA
GAAGAAGTCTTACAGGCTCTTCAGCTTCTTCACTGTTTCCCTCTGGACATACGATTAAAA
GATGGCAGTTTATTTTGGCAGTCACCAAAGAGGCCACCCTCTCCAATAAAATTTGATTTA
AATGAGCCTTTGCACCTCAGTTTCCTTCAGAATGCTGCAAAACTATATGCTACAGTATAT
TGTATTCCATTTGCAGAAGAGGTAAGATATTCT
>ENST00000505673 peptide: ENSP00000421984 pep:NOVEL_protein_coding
XCGAIGCEMLKNFALLGVGTSKEKGMITVTDPDLIEKSNLNRQFLFRPHHIQKPKSYTAA
DATLKINSQIKIDAHLNKVCPTTETIYNDEFYTKQDVIITALDNVEARRYVDSRCLANLR
PLLDSGTMGTKGHTEVIVPHLTESYNSHRDPPEEEIPFCTLKSFPAAIEHTIQWARDKFE
SSFSHKPSLFNKFWQTYSSAEEVLQALQLLHCFPLDIRLKDGSLFWQSPKRPPSPIKFDL
NEPLHLSFLQNAAKLYATVYCIPFAEEVRYS
`;

#Print out all of the sequences, if the sequence was DNA translate to Amino Acids
for @seqs -> $seq {
    say $seq ~~ BioInfo::Seq::DNA ?? $seq.translate !! $seq;
}

The previous example will output the protein sequence twice since the DNA was the coding sequence for the protein. Notice that what type of sequence was automatically detected, for very short sequences this could get confused. However, it does allow for convenient and quick manipualtion of sequences.

Installation

Installation is quite simple at the moment, but might become more involved once third party software is involved.

With zef (part of the Rakudo* distribution)

zef install BioInfo
perl6 -M BioInfo -e 'say BioInfo::Seq.new(id => "seq1", sequence => "MDADAFA");'

By hand

#Get the code into your library path
cd ~/wherever_you_keep_perl6_libs
git clone 'https://github.com/MattOates/BioInfo'
#Run some tests
cd BioInfo
prove -e 'perl6 -I ./lib' -lrv t/

License

Artistic, for now just so that I don't have to worry about hardcore Perl users being fearful of anything else.