Genomic DNA preparation enabling multiple replicate reads for accurate nanopore sequencing

Sequencing at single-nucleotide resolution using nanopore devices is performed with reported error rates 10.5-20.7% (Ip et al., 2015). Since errors occur randomly during sequencing, repeating the sequencing procedure for the same DNA strands several times can generate sequencing results based on consensus derived from replicate readings, thus reducing overall error rates. The method presented in this manuscript constructs copies of a nucleic acid molecule that are consecutively connected to the nucleic acid molecule. Such copies are useful because they can be sequenced by a nanopore device, enabling replicate reads, thus improving overall sequencing accuracy. Serial Copies Method Long-read sequencing technologies can be benefited significantly by sample preparation methods that enable multiple replicate reads of the same genomic DNA molecule. For example, the PacBio (Pacific Biosciences) sequencing error rate for a single read is relatively high (around 11%–15%) (Rhoads and Au, 2015). The Circular Consensus Sequencing (CCS) method allows for the repeated sequencing of individual templates (Travers et al., 2010). The errors are distributed randomly in single reads, so that the overall error rate can be reduced by generating CCS reads with sufficient sequencing passes. For example, templates with at least 4 replicate reads (i.e. templates that went through at least 4 CCS passes) have a minimum Phred quality score of 20, templates with 7 replicate reads have a Phred score of at least 40, and those with 9 replicate reads have a Phred score of at least 50 (Larsen et al., 2014). . CC-BY-NC-ND 4.0 International license peer-reviewed) is the author/funder. It is made available under a The copyright holder for this preprint (which was not . http://dx.doi.org/10.1101/035436 doi: bioRxiv preprint first posted online Dec. 28, 2015;