Error Correction in Nanopore Reads for de novo Genomic Assembly

The purpose of genome sequencing is to determine the DNA sequence of a given organism. Current sequencing technologies can be classified by the type of output data. Whereas Nanopore technology generates long reads with high error rates, short read technologies - such as Illumina sequencing - generate shorter reads but with low error rate. Since de novo genome assembly of sequencing reads is defined as a NP-hard problem, it remains as one of the major challenges for defining reference genomes of different species. This paper aims to improve the quality of reads obtained through Oxford Nanopore Technologies (ONT). We developed an algorithm to associate the reads obtained from Illumina with the ones obtained with Nanopore. Low accuracy ONT reads were corrected with the high quality Illumina reads to achieve an improved sequencing data. The inclusion of this algorithm as a preprocessing step resulted in improved coverage, contig length, and mismatch rate when performing de novo genome assembly of a bacterial genome with well known tools.