Mantas Sereika from Aalborg University in Germany presented at the Nanopore Community Meeting 2021 a session with a title that caught my attention tonight: “Nanopore R10.4 enables near-perfect bacterial genomes.” They spoke about Nanopore sequencing raw read accuracy improvements and issues with homopolymers. Insertions and deletions in homopolymer regions can be an issue causing frameshifts. Sereika used the R10.4 Q20 kit. The modal raw read accuracy was above Q20. Sereika and colleagues were interested in the performance of R10.4 Q20 performance on the Zymo mock community. Sereika hypothesized that the increased accuracy with Q20 and polishing with Medaka would lead to better assemblies. They tested this using the Zymo mock community sequenced at various coverages with and without Illumina read polishing. The results were encouraging with R10.4 sequences alone. Next, they assessed R10.4 Q20 performance on complex metagenome samples. Results indicated that there was an improvement in homopolymer calling of metagenome assembled genomes (MAGs). Nanopore R10.4 seems to be able to resolve hompolymers up to a length of ten. Further, with a coverage of forty and above, near-perfect genomes and MAGs can be obtained. Sereika was also excited about the impact of duplex reads approaching Q30. While duplex reads made up only about 1% of the throughput, Sereika explained that the read quality would be helpful in further improving bacterial genome assemblies. This session helped me better understand some of the issues with homopolymer sequencing and the impact of coverage.
