Figure 2

The NaS workflow. Inputs are the Illumina short reads and the MinION® reads (1D and 2D), purple bars on MinION® reads represent sequencing errors. Step1. Illumina reads are aligned on the MinION® templates to select seed-reads (light blue rectangles). Step2. Seed-reads are used to recruit similar reads in the initial Illumina read set. Step3. Good recruits (i.e., reads coming from the right genomic region) are light blue rectangles, bad recruits (i.e., reads coming from another similar genomic region) are red rectangles. Step4. Overlap-layout-consensus-based assembly of the recruited-reads and the seed-reads. Outputted contigs (light blue and red rectangles) are then filtered using seed-read alignments. In this example, after filtering step, a single contig representing the final NaS read is produced.