CSAM: Compressed SAM Format


Rodrigo Cánovas
L.I.R.M.M. & Institut Biologie Computationnelle, Universite de Montpellier, CNRS F-34392 Montpellier Cedex 5, France.

Alistair Moffat
Department of Computing and Information Systems The University of Melbourne, Victoria 3010, Australia.

Andrew Turpin
Department of Computing and Information Systems The University of Melbourne, Victoria 3010, Australia.


Status

Bioinformatics, 32(24):3709-3716, 2016.

Abstract

Motivation: Next generation sequencing machines produce vast amounts of genomic data. For the data to be useful, it is essential that it can be stored and manipulated efficiently. This work responds to the combined challenge of compressing genomic data, while providing fast access to regions of interest, without necessitating decompression of whole files.

Results: We describe CSAM (Compressed SAM format), a compression approach offering lossless and lossy compression for SAM files. The structures and techniques proposed are suitable for representing SAM files, as well as supporting fast access to the compressed information. They generate more compact lossless representations than BAM, which is currently the preferred lossless compressed SAM-equivalent format; and are self-contained, that is, they do not depend on any external resources to compress or decompress SAM files.

Availability: An implementation is available at https://github.com/rcanovas/libCSAM.


Full text

http://dx.doi.org/10.1093/bioinformatics/btw543.