Table 2.

Current challenges and solutions when using ONT to sequence plant genomes

Challenge	Potential solutions
Low DNA quality and quantity	Test multiple extraction protocols and optimize for each plant species.
Short read contamination	Removal of short and medium-sized fragments using BluePippin Prep or Circulomics Short Read Eliminator kits, the latter being easier to use.
Basecalling speed and computational requirements	PromethION includes the hardware needed for fast basecalling. MinION basecalling time can be significantly reduced by using GPUs.
Long assembly computation time	Newer assemblers can significantly reduce computational time (e.g. wtdbg2).
Remaining uncorrectable base errors	Additional Illumina sequencing and polishing is currently required (Watson and Warr, 2019). This might be addressed with newer pore versions or basecalling models trained for particular species. Useful software includes Racon and Pilon.
Assembly is not (near) chromosome scale	Additional techniques such as optical mapping or Hi-C can be used to order and place contigs and obtain (near) chromosome-scale assemblies, at least for small and medium-sized plant genomes.
Genome structural and functional annotation	For structural annotation, long-read technology can be used with programs such as Stringtie2 (Kovaka et al., 2019). For functional annotation, free online tools relying on specific plant expertise are available, such as Mercator (Schwacke et al., 2019), TRAPID (Van Bel et al., 2013), or Hayai (Ghelfi et al., 2019), in addition to general tools such as Blast2GO (Götz et al., 2008). The plant repeat database (Nussbaumer et al., 2013) can be used to analyse repetitive DNA, and structural variations can be analysed using NGMLR/sniffles (Sedlazeck et al., 2018).