Biopython and next generation sequencing

Those of you doing next generation sequencing may be pleased to know that the next release of Biopython is expected to include support for reading and writing FASTQ and QUAL files within our Bio.SeqIO interface. These formats are used for traditional Sanger capillary sequencing, and Roche 454 sequencing (Roche provide tools to convert from their binary SFF files) with PHRED quality scores. Solexa/Illumina sequencers produce a FASTQ variant where the quality scores are encoded differently, and this is also supported.

As part of this work, the SeqRecord has a new attribute holding a dictionary of per-letter-annotation (e.g. PHRED quality scores, Solexa quality scores, or for proteins secondary structure predictions), and the SeqRecord object can be sliced to give a new SeqRecord covering just part of the sequence, with the per-letter-annotation cropped to match.

The code is in our CVS tree, and mirrored on github, so those of you happy to install Biopython from source and help test it can grab the code now, and give it a go – feedback on the development mailing list please. Alternatively, sit back and await the forthcoming Biopython 1.50 (beta) release, which should allow more widespread testing…

Update: This new functionality was included in Biopython 1.50, including both Sanger standard FASTQ files and the original Solexa/Illumina FASTQ variant using Solexa scores. This did not cover the new FASTQ variant used on Solexa/Illumina 1.3+ but that is now supported in our development code, and will be included in Biopython 1.51. Peter, 2009/06/16


Categories