Extracting megabase DNA

Nadine Holmes from the University of Nottingham began by introducing the DeepSeq sequencing facility in Nottingham. They are a Nanopore certified service provider on GridION and PromethION. Continuing, Nadine explained why it’s important to sequence ultra-long reads: a) For improving assembly contiguity and phasing, b) resolving highly complex genomic regions and c) for detecting large structural variants. The lab has spent some time optimising ultra-long read sequencing methods -  pushing N50s and improving yields. The lab is also the current record holder for the longest read sequenced at 2.3Mb. To extract megabase DNA Nadine follows Josh Quick's method. This method can sometimes yield DNA concentrations exceeding 5μg /μl. For QC Nadine suggested it is best to dilute the DNA to 1μg/μl to make easier to handle. Nadine stressed the importance of DNA quality control. She said that if a femtopulse or PGFE is not available a tapestation still works well to identify badly fragmented samples. Nadine also uses the Pippin pulse to look at fragment lengths of HMW DNA.

Nadine continued to explain the method for generating megabase reads. 15ug of DNA is used in the RAD004 kit following Josh Quick's published method. Mixing reagents and DNA very slowly with wide bore tips is important, library loading beads are omitted  and  the final library is run with active channel selection turned off. To date the Loose lab has run >40 flow cells with this ultra-long read method and it typically generates 1-2Gb per MinION flow cell with N50’s from 50 to > 100Kb. Nadine mentioned that yield and N50 can be extremely variable and are dependent on the sample. Nadine continued to discuss how the recording breaking 2.3Mb read had initially been split up by MinKNOW. Using the labs Bulk-viz software package they were able to identify the issue, join the reads back together and claim the long read title. Nadine then spoke briefly on their success on translating the method onto flongle and mentioned that optimisation work is ongoing to scale this up onto PromethION flow cells.

Breifly In the last part of Nadine’s talk she spoke about her experience with some commercial kits having good success with SageHLS and Circulomics Nanobind extraction methods. Nadine then spoke highly of the Circulomics short read eliminator kit for removing short fragments from poor customer samples the lab had received.

Finishing up her talk Nadine said it was crucial to optimise the extraction method for the project in hand. Some very light shearing before running the 1D ligation prep can help to get good yields. She has also had really good success with size selection methods (blue pippin and circulomics), and good success with the Nanopore nuclease flush protocol.