Admitting that it’s a bit “unusual” for a software development project to take almost eight years to reach its first whole-number release, Jason Stajich said last week’s launch of BioPerl 1.0 was a long time in coming because its developers simply wanted to get it right.
“We really wanted to be confident that what we were promising was really there,” said Stajich, a PhD candidate at Duke University who heads up the project.
Stajich admitted that the 1.0 stamp, while genuinely denoting a fully stable and “feature complete” toolkit for the majority of biological sequence manipulations in bioinformatics, is “cosmetic in a way.” While there had been a “certain amount of pressure” to put out a 1.0 release, Stajich said a high level of functionality was there all along. “The entire human genome was annotated on BioPerl 0.7,” he noted.
According to its developers, the 1.0 release contains modules for representing biological sequences, protein structure, sequence alignments, Blast and Fasta reports, biological maps, sequence features and their locations, annotations and bibliographic references, phylogenetic trees, and gene structures. It also includes tools to visually render sequences, build and access local sequence indices, and parse and output most major biological sequence formats. Stajich said the toolkit has been made more user-friendly as well, and additional documentation has been added to assist new users.
James Diggans, a bioinformatics programmer at Gene Logic who uses BioPerl, said the difference “is night and day” between 1.0 and the last stable release, 0.7. “The parsing model for Blast reports has gone to an event-based model, which is much faster and more memory efficient,” he said. “It’s a pretty robust solution.”
The major advantage of the toolkit, according to Diggans, is that it offers a common set of solutions for routine bioinformatics tasks. “If you have a bug in one application, you don’t have to fix it a hundred times. It gets fixed in one place and everyone benefits from the solution.”
Stajich said the hope is that bioinformatics developers will “recognize BioPerl as an accepted standard” and turn to the toolkit instead of duplicating the Perl scripts for the same tasks in different organizations.
BioPerl 1.0 is available at www.bioperl.org.