Software Background

Canu is derived from Celera Assembler, which is no longer maintained.

Celera Assembler [Myers 2000] was designed to reconstruct mammalian chromosomal DNA sequences from the short fragments of a whole genome shotgun sequencing project. Celera Assembler was used to produce reconstructions of several large genomes, namely those of Homo sapiens [Venter 2001], Mus musculus [Mural 2002], Rattus norvegicus [unpublished data], Canis familiaris [Kirkness 2003], Drosophila melanogaster [Adams 2000], and Anopheles gambiae [Holt 2001]. Celera Assembler was shown to be very accurate when its reconstruction of the human genome was compared to independent reconstructions completed later [Istrail 2004]. It was used to reconstructing one of the first large-scale metagenomic projects [Venter 2004, Rusch 2007] and a diploid human reference [Levy 2007, Denisov 2008]. It was adapted to 454 Pyrosequencing [Miller 2008] and PacBio sequencing [Koren 2012], demonstrating finished bacterial genomes [Koren 2013] and efficient algorithms for eukaryotic assembly [Berlin 2015].

Celera Assembler was released under the GNU General Public License, version 2 as a supplement to [Istrail 2004].

Canu [Koren and Walenz 2017] was branched from Celera Assembler in 2015, and specialized for single-molecule high-noise sequences. For the most recent license information on Canu, see README.licences.