Editing Portal:Life Sciences (section)

=== Why biology needed Perl ===

Molecular biology produces data that looks simple but is computationally demanding. DNA and RNA sequences are strings over small alphabets. Protein sequences are strings over amino-acid alphabets. Sequence annotations, database records, alignment files, BLAST reports, FASTA files, GenBank flat files, GFF files, quality scores, laboratory metadata, and instrument output are all text-heavy and format-heavy.

This made Perl unusually useful. Perl’s regular expressions, arrays, hashes, file handling, string manipulation, and Unix integration were a natural match for biological data. A Perl script could read a FASTA file, extract identifiers, clean sequence names, run an external alignment program, parse the output, load results into a database, and generate a report.

Perl became popular in bioinformatics because it solved immediate problems. Researchers did not need a large software-engineering framework to begin. They could write a script, test it on a few sequences, run it on a directory of files, and then expand it into a pipeline.

Common Perl bioinformatics tasks included:

* parsing FASTA, GenBank, EMBL, SwissProt, GFF, BED, SAM-like, and tabular files;
* cleaning sequence identifiers and laboratory sample names;
* extracting genes, exons, introns, coding regions, and protein translations;
* running BLAST and parsing BLAST output;
* filtering alignments;
* converting between file formats;
* building genome-annotation reports;
* connecting sequence files to relational databases;
* automating genome-center workflows;
* generating web pages and browser tracks;
* coordinating large batches of command-line tools.

This style of work made Perl the “duct tape” of early genomics.