Entering edit mode
4.7 years ago
kamanovae
▴
100
Hello everyone, I want to sequence the human genome using Oxford Nanopore Technology. I came across various papers of ONT in the year 2019 on this topic and in all studies use 1D protocols technology, not 1D2.
I listed the most popular articles here: 1) https://www.nature.com/articles/nbt.4060#methods 2) https://www.nature.com/articles/s41467-019-09637-5#Sec9
Could anybody recommend me what library preparation techniques should I use for sequencing the full human genome?
Probably by the time they did the experiments 1D2 was not available. If something got published in January 2018 (the first link) you have to subtract some month from submission of the manuscript over revision to final publication etc. plus performing and analyzing the experiments etc, can easily be some month/years. The 1D2 was released middle 2017 I think, so little chance they already used it/ were aware of it at that time.
Plus you need to add time for the community to evaluate how the new technology performs in the wild and how much more the new technology costs vs its benefits.
The 1D² basecaller is guessing which of the reads are pairs (meaning one read is the complement of the other). Reads might get matched erroneously. If you have access to the ONT community forums there is a post explaining this problem in depth. I have heard people say they feel 1D² is just not worth it, they didn't get better results compared to the standard 1D chemistry.
Personally, i have only gotten my hands on one single 1D² dataset. So not much to go on. But in this one dataset the 1D² reads were horrible, not even worth the time it took to basecall them.
It also feels like ONT themselves aren't really betting on the 1D² method. Take a look at ONTs own medaka, their new (and pretty awesome) polishing software. It has a range of models, even one for the old transducer algorithm. But no model for the R9.5 pore (the one used on the 1D² flowcells).