A de novo DNA Sequencing and Variant Calling Algorithm for Nanopores

The single-molecule accuracy of nanopore sequencing has been an area of rapid academic and commercial advancement, but remains insufficient for the de novo analysis of genomes. We introduce here a novel algorithm for the error correction of nanopore data, utilizing statistical models of the physical system in order to obtain high accuracy de novo sequences at a range of coverage depths. We demonstrate the technique by sequencing M13 bacteriophage DNA to 99% accuracy at moderate coverage as well as its use in an assembly pipeline by sequencing {lambda} DNA at a range of coverages. We also show the algorithms ability to accurately classify sequence variants at far lower coverage than existing methods.