Considerations and Performance Data for Nanopore Sequencing of
High Molecular Weight DNA (HMW DNA)

Successful nanopore sequencing requires standard molecular biology processing including DNA repair, end-repair, adapter ligation, and cleanup. As such, the fragment length of material used is an important consideration during experiment design. According to internal testing, the use of the standard ligation-based Oxford Nanopore Technologies® protocol with agitation at 2,000 rpm during lysis produced the highest N50 values, maximal percent of active pores on the flow cell, and maximal amount of data per run (often exceeding 10 Gb of data without flushing the flow cell). Using lower agitation speeds for longer DNA may seem logical since longer fragments should theoretically produce longer contigs. However, library prep of HMW DNA above a certain size has been shown to provide inferior results. It is recommended to target a DNA size range of 50–250 kb for optimal sequencing results. 

Following library prep, consider loading additional DNA library onto the flow cell. Because the size of the DNA molecules extracted with this method are so long, more input material is needed to maintain the proper molar ratios. Loading 800–1500 ng of HMW gDNA on the flow cell provides optimal sequencing results in NEB’s internal testing. 


Single Run Sequencing Results of HEK293, Human Blood, and Mouse Kidney Samples on the Oxford Nanopore Technologies Platform 

  HEK293 SAMPLE 1 HEK293 SAMPLE 2 HUMAN BLOOD SAMPLE 1 HUMAN BLOOD SAMPLE 2 MOUSE KIDNEY SAMPLE
Mean read length (bases) 21338.9 19249.9 21522.6 24677.7 27120.7
Mean read quality 12.8 13.2 13.4 13.3 13
Median read length 10388 9702 10130 12593 23150
Median read quality 13.2 13.7 13.9 13.8 13.5
Number of reads 377687 633636 538090 327314 164000
Read length N50 (bases) 45432 40415 46542 51394 44631
Total bases 8059414490
(8.1 Gb)
12197410796
(12.2 Gb)
11581090785
(11.6 Gb)

8077351338
(8.1 Gb)

4447789727
(4.4 Gb)
Number, percentage and megabases of reads above quality cutoffs
> Q5 377687 (100.0%)
8059.4 Mb
633636 (100.0%)
12197.4 Mb
538090 (100.0%)
11581.1 Mb
327314 (100.0%)
8077.4 Mb
164000 (100.0%)
4447.8 Mb
> Q7 377685 (100.0%)
8059.3 Mb
633636 (100.0%)
12197.4 Mb
538090 (100.0%)
11581.1 Mb
327314 (100.0%)
8077.4 Mb
163999 (100.0%)
4447.8 Mb
> Q10 346678 (91.8%)
7546.9 Mb
578938 (91.4%)
11333.7 Mb
498562 (92.7%)
10864.5 Mb
298205 (91.1%)
7496.1 Mb
148584 (90.6%)
4099.9 Mb
> Q12 278174 (73.7%)
6218.2 Mb
484738 (76.5%)
9725.8 Mb
425604 (79.1%)
9438.5 Mb
253520 (77.5%)
6495.0 Mb
124810 (76.1%)
3526.1 Mb
> Q15 22058 (5.8%)
333.2 Mb
109840 (17.3%)
1986.8 Mb
113362 (21.1%)
2348.0 Mb
60357 (18.4%)
1371.8 Mb
12964 (7.9%)
305.3 Mb
bottom 5 highest mean basecall quality scores and their read lengths
1:00 19.1 (332) 19.9 (263) 20.6 (156) 19.6 (332) 18.7 (229)
2:00 18.6 (268) 19.8 (142) 19.9 (214) 19.2 (1043) 18.7 (1041)
3:00 18.5 (319) 19.8 (202) 19.5 (634) 19.2 (1317) 18.5 (792)
4:00 18.3 (645) 19.6 (631) 19.2 (366) 19.2 (570) 18.4 (200)
5:00 18.2 (1500) 19.2 (192) 19.1 (4073) 19.1 (303) 18.4 (550)
bottom 5 longest reads and their mean basecall quality score
1:00 441385 (13.0) 277445 (12.1) 298409 (12.3) 333983 (13.1) 200926 (12.3)
2:00 331772 (12.8) 249785 (11.1) 282024 (13.3) 291973 (13.3) 190512 (11.2)
3:00 299267 (13.2) 248331 (11.5) 280729 (14.1) 278424 (12.5) 182388 (14.5)
4:00 292461 (13.1) 246374 (14.8) 275886 (12.5) 274430 (11.3) 181646 (9.2)
5:00 272042 (13.8) 244162 (13.5) 271731 (12.1) 273123 (12.2) HMW 177197 (13.0)

DNA used for the sequencing libraries was extracted by following the standard protocols for cultured cells and fresh mammalian blood samples, without further size selection. Mouse kidney DNA samples were extracted from fresh samples using a rotor-stator homogenizer for sample homogenization. Libraries were generated using the LSK109 ligation sequencing kit and loaded on a FLO-MIN106D flow cell. Sequencing was performed on a GridION® Mk1 for up to 48 hours, or shorter if no more data was generated by the flow cell. No additional treatment of the flow cell (e.g., flushing) was employed. In the samples that resulted in > 10 Gb of data, 800–1,000 ng of DNA library was loaded onto the flow cell for optimal sequencing performance and effective pore usage.&




Related Products:

Related Resources: